The phrase “it’s coming home” has, over the years, taken on a hint of irony for English football fans. After all, it’s been six decades since England last won a major international trophy — a drought that rivals just about any in sports.

But this year has been pretty good for ending trophy droughts: Arsenal won their first Premier League in twenty-two years — the first since their famous Invincibles season — and the Knicks snapped a 53-year wait, beating the Spurs for their first NBA title since 1973.

On the surface, the English national football team is a lot like the Knicks: both believe they’re the rightful “home” of their respective sports, both ran into juggernauts in the late 1990s and early aughts, both suffered through bad management in the early 2010s, and both still have some of the most passionate fans — maybe too passionate — in the world. By now the relationship between English fans and their team borders on tragic, each tournament following the same script: a wave of genuine belief, then the familiar blow — a penalty shootout or quarterfinal collapse.

Despite their trophy drought, England remains a hotbed for producing world-class players.1 After they laid waste to Luka Modric and Croatia in the second half of their opening game, it was easy to see the type of performances their squad is capable of. Still, while England certainly has the talent to succeed this year, far more goes into winning the World Cup than having the best players.

Subscribe now

Why did England’s Golden Generation fail?

A lack of talent has never been the real issue for England. In fact, too much of it may have been the root of their underachievement in the 2000s. During the 2000s, the English talent wave — nicknamed its country’s Golden Generation — was deep at every position.

They boasted two monstrous attackers in Michael Owen and Wayne Rooney. Owen won the Ballon d’or in 2001 at only 22 years old, while Rooney is recognized as one of England’s finest-ever talents, starring for Manchester United teams that dominated the Premier League. Their backline had no holes either: Rio Ferdinand is widely considered one of the best defenders of his generation, Ashley Cole and Gary Neville flanked the left and right back positions, respectively, and John Terry filled in the final spot. Their midfield was perhaps their deepest position: Joe Cole, Paul Scholes, Steven Gerrard, Frank Lampard, and their captain, David Beckham, competed for scrace available spots in the center, often leaving an odd man out.

England’s Golden Generation was something of a cultural phenomenon at the height of tabloid culture, with the footballers’ relationships being some of the most discussed news stories in mid-2000s Britain, peaking around the 2006 World Cup in Germany. It wasn't without reason: the English squad boasted some of the most recognizable footballers on the planet, but their partners — or as British media coined them, “WAGs” — were famous in their own right. Household names like Victoria Beckham and Cheryl Cole made their partners into news fixtures as much as any goal or trophy did.

These relationships were engines of financial success. David Beckham was a wildly good footballer at the peak of his powers, yes, but the commercial value of his brand proved to be far more valuable to clubs looking to build viewership. At the time of the 2006 World Cup, Beckham was playing for Real Madrid as one of their Galacticos, but was already considering offers to play in the MLS as part of a plan to build the global fame of the league.

Despite the stardom, the Golden Generation didn’t produce a single trophy, never making it further than the quarterfinals in any of the major tournaments during the 2000s.

Of course, there are some boring reasons behind England’s underachievement in major tournaments. For one, there was quite a lot of bad luck involved. England had a longstanding “penalty curse.” Across a decade of major tournaments, they were repeatedly undone from the spot: Euro 96 against Germany, a 1998 World Cup last-16 tie against Argentina, Euro 2004 against Portugal, and the 2006 World Cup quarterfinal against Portugal again. England also faced major injuries during some of these tournaments: Michael Owen famously tore his ACL during the 2006 World Cup, and Wayne Rooney struggled with metatarsal injuries after breaking the bone in the lead-up to that same tournament.

Even then, not reaching even one semifinal for almost a full decade can’t be explained only by bad luck. England had its fair share of poor performances, highlighted by poor chemistry between their squad and rigid tactical formations that didn’t showcase their best players.

Manager Sven-Goran Eriksson employed a 4-4-2, which put Gerrard and Lampard — both gifted attacking midfielders — at odds with one another. The fundamental issue was that the 4-4-2 asks its two central midfielders to split duties, with one sitting back to provide defensive cover while the other roams forward to create. Gerrard and Lampard were both creators, by temperament and by the role they played at club level. When both pushed forward, as they tended to do, they vacated the space in front of the back four and left England exposed through the middle on transition. When one held back, England no longer retained the midfield advantage it had on paper.

England could have employed defensive midfielder Owen Hargreaves in a more prominent role, but he often didn’t get the call to start. English management had trouble leaving a world-class player like Lampard, Gerrard, or Scholes on the bench, and Hargreaves was often derided as England’s lesser midfielding talent. Hargreaves turned out to be one of the few bright spots in an otherwise underwhelming 2006 campaign, proving the media wrong, though it didn’t rescue England's tournament.

Some of England's misfortune, too, was simply a matter of being one superteam among many in a crowded era. In 2002, the Brazilians featured Ronaldo, Ronaldinho, and Rivaldo, and beat England in the quarterfinals before winning the tournament in dominant fashion. In 2006, the Italian side that won the tournament featured world-class players up and down the pitch, from Gianluigi Buffon in goal to Ballon d’Or winner Fabio Cannavaro in defense and Andrea Pirlo pulling the strings in midfield. England’s trouble against a strong Portuguese side in 2004 and 2006 shouldn’t be excused, but Portugal wasn’t exactly an underdog. After all, they started Luis Figo, Deco, and a young Cristiano Ronaldo in a devastating attacking trio that would have troubled any good team.

Even in their better moments, England never achieved a year-end #1 PELE ranking, and only ever peaked at year-end #4 during Beckham’s run as team captain. Moreover, the dominance of countries like Spain highlighted England’s tactical rigidity. During the 2008 Euros, Spain ran a more typical 4-4-2 formation, but by 2010, they had started to focus more on possession as the main lever of control, treating the ball itself as the most valuable thing on the pitch and building everything else around keeping it.

Under Luis Aragones and then Vicente del Bosque, Spain moved from a system of football nicknamed la furia — “Spanish fury,” an identity built on physicality and aggression — to one known as tiki-taka, a possession-obsessed style defined by short passing and constant movement. In the 2012 Euros final, for example, Spain lined up without a recognized striker at all, deploying the midfielder Cesc Fabregas as a false nine ahead of a side built around Xavi, Busquets, Xabi Alonso, Iniesta, and David Silva in a now-famous 4-6-0 formation. This stylistic evolution underpinned Spanish play during a period in which many other footballing powers struggled to get the most out of their talent.

Of course, you can't blame England for not pushing the boundaries of football any more than you can blame early 2000s NBA teams for not shooting enough threes. Innovation involves risk-taking, and the number of chances you have to win an international trophy with a Golden Generation of players is few and far between.

Can England bring it home this time?

The modern England side has at least somewhat internalized the lesson the Golden Generation never could. Under new manager Thomas Tuchel, who took charge at the start of 2025, the team has a clear division of labor in midfield, and a willingness to leave big names at home when they don't fit.

The 2026 iteration had some huge omissions — Phil Foden, Trent Alexander-Arnold, and Cole Palmer all missed out after mediocre club seasons —and some in the press have called it the most shocking English squad in almost thirty years.

These decisive moves have paid off for England thus far. Despite a draw against Ghana in their last match, they’ve enjoyed a stellar run of play going into this World Cup. They won all eight qualifiers without conceding a single goal, and outplayed familiar foe Croatia in their opening match.

Kane, unlike the Golden Generation players, is a bit underrated considering his consistency over the last decade or so. Most serious football analysts consider him the most complete striker in the world, even if fans don’t always give him his respect: he’s England’s all-time leading scorer and has maintained form in both the Premier League and the Bundesliga.

When Kane first burst onto the scene in Tottenham, he was considered more of a poacher — reliable, yes, but not a player that England could anchor a counterattack around. By 2021, Kane had started to use his attacking potential to draw defenders and create space for fellow Tottenham left-winger Son Heung-Min, which led to a devastating number of goal-combinations.

This expanded framework for the #9 role has become one of the most devastating aspects of Kane’s game. During his 2017-18 season with Tottenham, Kane operated as a penalty-box striker, but he’s gradually moved away from poaching to a deep-lying creator role, playing closer to the halfway line. Despite many predicting Kane would have a steep decline in older age, he’s effectively tripled his creative output, creating more for his teammates.

Kane isn’t necessarily unique in this transformation. Lionel Messi transformed from a right-winger to a deep-lying playmaker later in his career, going from a devastating one-on-one dribbler during his early years at Barcelona to a game built on vision and weight of pass. It’s the appropriate way to transition your game from being dependent on athletic tools to something more stable: intelligence and positioning.

The shift has steadied Kane as one of the world’s best players— there are arguments he is outright the best, though I won’t go too far — but England should be wary of an overreliance on him. So far, Tuchel has been keen to make use of Kane’s all-around offensive versatility, but their dependence on him could put them in tight spots as they go deeper into the knockout rounds.

He’s produced nearly a third of the team’s goals and xG across the last four tournaments, but when he struggles to break down the opposition’s backline, they can be left vulnerable to tight games that get decided by small margins. In England's match against Ghana, for example, Kane said he couldn't find much space because he was effectively man-marked by Thomas Partey, denying him the room to drop deep and arrive late in the box. England dominated possession (79 percent) and outshot Ghana 19-2, but were essentially neutralized because there wasn’t a second reliable attacker for Kane to create space for.

Nonetheless, Kane isn’t the only world-class talent on their roster. England’s most pivotal player might be Arsenal midfielder Declan Rice, widely regarded as one of the most complete players in football. Rice’s transformation from playing the destroyer role to becoming a more versatile box-to-box attacker catapulted Arsenal to the Premier League title and a Champions League final. The unlock came when Arsenal signed Spanish holding midfielder Martín Zubimendi in 2025. With Zubimendi anchoring the base of midfield in a double pivot, Arteta was free to push Rice into a more advanced role, releasing him from the pure shielding duties that had defined his West Ham years. Rice topped Arsenal’s rankings for both chances created and possession won, and even with 27 of his 63 key passes coming from set pieces, he still finished joint-second on the team for chances created in open play while remaining their primary ball-winner.

Tuchel has kept him in that role thus far, pushing him relatively far above the halfway line in England’s first two matches. If Rice’s presence can aid England’s attack and relieve some pressure off Kane and fellow midfielder Jude Bellingham, England looks to be extremely powerful in this tournament.

So will England bring it home this year? They have a good chance — our model gives them the fifth best odds in the field — but it might be worth tempering the optimism. For one, England will need to win today against Panama. They’ve already clinched a spot in the knockouts, but finishing second could mean playing either one of Portugal or Colombia in the round of 32. Both teams are good enough to send England home. There’s a world in which they could play a very tricky combination of opponents in sequence: Portugal, Spain, Belgium and France — and winning out would require both some luck and tactical brilliance. Conversely, a win today could put them on the other side of the bracket, where they would play DR Congo or Senegal in the first round. Senegal is the better team, but is probably England’s preferred matchup, given their defensive limitations thus far.

Either way, World Cup title or not, England’s tactical innovations are suggestive of a side that has gotten over its tendency to favor star players over the overall healthiness of their team. Much of winning a World Cup is luck, but after 60 years, a team built to give itself the best possible chance is about all you can ask for.

After years of build-up, the Giannis Antetokounmpo trade saga drew to a close Monday night, with the Bucks’ franchise legend heading to Miami. Per various reports, Milwaukee spent the afternoon weighing two offers: a Celtics package fronted by some variation of Jaylen Brown and draft capital, and the Heat bid that ultimately won out.

They ended up shipping Antetokounmpo and Bobby Portis for Tyler Herro, Kel’el Ware, Jaime Jaquez Jr., Kasparas Jakucionis, three first-round picks, a pick swap, and a second-rounder. Jon Horst and the rest of the front office chose flexibility over a win-now star, taking a stockpile of capital to build the next era of Bucks basketball around upcoming talent rather than an established veteran.

There are more than a few angles to discuss here. But the trade also feels like an anticlimax, as Giannis’s star has dimmed from consensus top-3 player to a guy who might need to be paired with another star or an outstanding core of complementary players to compete for a title.

Subscribe now

Milwaukee fumbled a golden opportunity

It’d be hard not to feel a little disappointed if you’re a Bucks fan. None of the players they received have real All-NBA upside, and the pick they acquired in this year’s draft landed outside the top ten, leaving them with one of the draft’s more polarizing players in Nate Ament. They did pick up Miami’s 2031 and 2033 first-rounders, which could become valuable assets down the line, especially with the league’s new lottery system.

Trading players when they can fetch the most value is optimal portfolio management — though “selling high” is difficult for any number of reasons. Still, the Bucks getting cold feet more than a few times over the last few years has cost them some potential assets. For what it’s worth, despite my own projection that Giannis would regress, he looked very good in the half-season that he played this year — but not quite good enough for his value as an asset to catch up to his age and injury concerns.

While not everyone inherently trusts the advanced stats, Giannis’ defensive EPM tracks with consensus opinion here. He won Defensive Player of the Year in 2020, also his best season according to EPM. But he hasn’t made an All-Defensive team since 2022. The pattern is typical of aging players, who sometimes hide the decline with less effort on the defensive end while maintaining the same “box score stats”. It’s not much of a decline on a per-minute basis, though, and any player with Giannis’s geometry is going to be severely disruptive at the rim. The question is availability.

Giannis is also on the last year of an expiring contract with a player option — he’ll almost certainly opt out and fetch more money — which gave him leverage on his trade destination. According to reports, he narrowed his own preferences down to Boston, Miami, and strangely, Minnesota (which wound up having other ideas up its sleeve.) None of those teams could put together the world-beating packages you would expect for a top five player in the league.

Still, the absence of asset-rich teams like the Spurs, Rockets and Thunder from the conversation should be noted. If the mock trades from a year ago were anything close to Giannis’ true value in the moment, the Bucks missed out on a chance to land a real cornerstone. Most of the Bucks’ conversations stayed behind closed doors, so the rumored packages deserve a healthy dose of scrutiny. Still, if there was a deal on the table headlined by a genuine young building block, like Stephon Castle, plus the usual draft capital, it would exceed what they ended up with this year.

Or there was another possibility: the Mavericks reportedly offered Luka Doncic for Giannis. Perhaps that rumor was a bit of ass-covering for Nico Harrison, whose process in trading Luka to the Lakers was secretive. But let’s say they made that trade. It’s possible Luka wouldn’t have wanted to stick in Milwaukee long-term. They’d still have held his Bird rights and been able to offer the most years and money of any team, so he’d likely have been in a sign-and-trade of some sort. Could they have fetched a better package for him than they just received from the Heat? Both Luka and Giannis have injury concerns and fit issues with teams not specifically designed around them, but Luka is four years younger, and is probably the better asset as of this moment both in theory and in practice.

By now some of our Wisconsin readers — there are 2,763 of you — are probably itching to change the subject. Forget about what the Bucks should have done. What happens now? Well, the trade package isn’t so bad. As I mentioned, the Heat attached their 2031 and 2033 first rounders, which are more valuable under the new lottery system (teams can no longer just win enough games to avoid conceding a good pick), and the young talent they received is enough to be tentatively excited about. They retain Ryan Rollins, who made a sneaky run at the Most Improved Player award last season, and Kel’el Ware is an intriguing player who was buried on the Heat’s bench for unclear reasons.

Granted, if you're a Bucks fan, supporting a team that's openly rebuilding without control of its own future first-round picks might feel a little strange. There are ways out of it, but they'll take a kind of patience you may not be used to. The last time the Bucks missed the playoffs for two seasons in a row was actually before Giannis got drafted. Barring any further moves, it’s likely that streak is broken this upcoming season.

Miami gets the better of Boston again — or did they?

After their many playoff battles, it was only fitting that Miami foiled Boston’s plans to retake control of the Eastern Conference. It’s hard to say, though, that either team is in a significantly better place after the trade had settled.

On the surface, Miami’s roster doesn’t suggest championship contention. The pairing of Bam and Giannis is fine, on both ends, but the rest of Miami’s roster has been stripped down to its bones. They can try to resign Norman Powell on a cheaper deal — it looks unlikely at the moment — and fill in the rest of their roster with minimum contracts, but does that team compete with the Knicks, Celtics, Pistons, or even the Pacers1 next season?

I’m inclined to say no, but that doesn’t mean that they overpaid. Some of my favorite pieces to write at Silver Bulletin are our semi-annual Future of the Franchise rankings, particularly because they help us frame transactions in terms of championship equity. Does Giannis give the Heat enough championship equity to justify mortgaging their future? It’s certainly possible. Something I’ve learned to not underrate is the value of short-term championship equity.

The Heat could lose in the playoffs next season, but all they really need is one or two strong runs over the next four years, barring major injuries to Giannis or Bam. And the league’s Collective Bargaining Agreement is on their side here: the hardest thing to acquire in today’s NBA is a cornerstone, not the role players around one. It’s more or less impossible to sign your way to a top-five player, while rotation pieces, by contrast, are a renewable resource — minimums, the mid-level, veterans chasing a ring all present affordable options.

The bull case for Miami doesn’t rest on this roster beating New York next May. It’s that they’ve already cleared the highest hurdle and the CBA makes everything downstream of that much cheaper to solve for a front office in a desirable market that tends to make strong moves on the margin.

Still, the clock is ticking. “Barring injuries” is not something to just gloss over — after all, it was baked into Giannis’ trade value. The trade reminds me a bit of the Kevin Garnett sweepstakes in 2007 — the Celtics traded a lot of their middling depth for the aging superstar and ended up winning a championship, with Garnett playing 71 games for the terrific 2008 team. By 2009, Garnett was already dealing with major knee injuries that would speed up his decline. If the “Big Three” hadn’t won a ring together, would the trade be considered as much of a win? Still, that’s somewhere in the vicinity of a favorable precedent for this deal.

Of course, Giannis has proven to be more durable in form than Garnett was at the same age, but he hasn’t finished a season truly healthy since 2022. This isn’t very surprising for a player over the age of thirty.

Heat fans might have to adjust their expectations. This team probably isn’t going to compete for the #1 seed even if they have decent health next season.

Still, it’s worth noting that the Heat often outplay the sum of their parts. With Erik Spoelstra on their bench, and a group of scrappy veterans, it’s not impossible they can make some noise next year, even if they don’t end up winning the championship. If you’re a Heat fan, that would obviously be a better outcome than the play-in disappointments of the last few years.

The Jaylen Brown sweepstakes begin?

Even if the Giannis trade makes the Heat competitive again, the rest of the league should be breathing a collective sigh of relief that the Celtics couldn't pull off their attempted heist of the Bucks. I know Jaylen Brown just made an All-NBA team, but swapping him and Giannis for only two additional picks would have been highway robbery.

Had the trade gone through, the Eastern Conference landscape would have shifted drastically in favor of the Celtics, especially given the depth they have to surround a star like Giannis. Luckily for the rest of the league, the Celtics are now in a somewhat weird spot — Brown can’t be happy that Boston so openly tried to offload him after a career year, and it looks like they’ll be taking calls for him.

I don’t love to speculate on future transactions, but I’d be surprised if he’s still on Boston’s roster to start next season. The market for Brown is sort of ambiguous at the moment, though. It seems like Milwaukee’s biggest concern in trading for him was a worry that he would demand a trade elsewhere, not that he wasn’t an appropriate talent return, so there may be interest around the league.

If Brown really does have a top fifteen trade value, like Bill Simmons ranked him back in February, then the offers should be steep. Depending on the front office making the trade, it seems plausible he could be valued at a very high price, even if most of the advanced metrics consider him as a solid starter at best. Brown had a career year, sure, but inflated per-game numbers without a matching jump in the advanced metrics is about as clean a sell-high window as you’ll find.

Why is Brown not liked by impact metrics? Well, he’s a fine scorer on high volume, but his passing doesn’t match his overall offensive load while he averages more turnovers than you would expect for someone of his archetype. Even with his scoring getting directionally better, there are clear ceilings on what he can do in a lead role. On defense, he has the reputation as a two-way player but has regressed into more of a one-way, efficient-but-not-super-efficient shot creator. Of course, trade value and impact metrics can differ. Shot creators are scarce in the NBA, especially after Brown demonstrated he could carry a high usage rate. Ultimately, I would expect Brown to be moved for at least a credible package of capital — at least two firsts and/or a younger player with upside — even if his impact estimate doesn’t justify it.

The Timberwolves just made the best move of the summer so far

The Giannis trade dominated the news cycle, but a few moves flew under the radar this week, most of all the Timberwolves’ offloading of Julius Randle. The trade was made to open up financial flexibility for an Ayo Dosunmu extension — a pretty large overpay in my opinion — but left room for further moves. It was rumored that the Timberwolves tried to trade for Derrick White, but were turned down shortly before the Giannis negotiations hit their peak.

As it turns out, those further moves came fast. With rumblings about Anthony Edwards’ reported unhappiness clearly weighing on the front office, Minnesota pulled the trigger on the real blockbuster: LaMelo Ball (plus Josh Green) for Naz Reid, a 2033 unprotected first, three first-round swaps, and a bundle of seconds. As a LaMelo Ball fan (although I can admit he has some flaws, both as a player and an off-court personality), I think this is an very solid gamble for the Wolves — and a case where it pays to be long volatility given the high bar established by the Spurs and Thunder in the West.

It’s a weird move for Charlotte. LaMelo is the engine of their offense, and the Hornets were one of the best teams in the league over the second half of the season, but perhaps they have doubts about his injury history or his place as a franchise cornerstone.

Still, the return isn’t amazing when compared to LaMelo’s overall impact2. It’s not certain that any of those swaps would convey — the Hornets are probably going to be worse than the Timberwolves for a while — and Naz Reid is a fine player, but nowhere close to LaMelo as a ceiling raiser. The bet is mostly that either LaMelo will be injured or Anthony Edwards demands a trade, which would let one of those swaps actually land in the lottery. I feel this isn’t a basketball bet so much as a bet against Minnesota’s stability, which could be plus-EV in terms of pure championship equity (assuming LaMelo can’t recreate his past season elsewhere) but additional years of struggle for a franchise that last made the playoffs a decade ago.

The rest of the league is still reacting to this past season’s playoffs, which appear to have created something of an arm’s-race as teams adjust to the talent level now required to compete for a championship.3 Most front offices haven’t fully tipped their hand yet, but the Giannis and LaMelo deals suggest that more dominoes will fall in the coming weeks, and that the order in both conferences could look fairly different by the time they do.

It’s been a terrific World Cup — but the tournament will only accelerate from here, with the highest-stakes matches yet to come. Teams will play their third and final group-stage matches in simultaneous pairs; the structure is intended to reduce the informational advantage from knowing exactly what you’re playing for.1 Then we’ll have the first-ever 32-team knockout bracket.

This is going to be an extremely meat-and-potatoes article. (Thanks to the thousands (!) of you who have subscribed for our World Cup coverage.) I’m simply going to go through each of the 12 groups and describe what each of the 48 teams has at stake in its final match. Only 7 teams have officially advanced to the knockout round — these are, in alphabetical order: Argentina, Colombia, France, Germany, Mexico, Norway and the United States, each with 6 points.2 Not all of these teams have clinched their groups, however. Another 11 teams are sitting on 4 points, with a win and a draw so far. They are extremely likely to advance, although in some cases, whether they finish 1st, 2nd, or 3rd in their group makes a big difference.

Only 5 teams have been formally eliminated: Haiti, Jordan, Panama, Tunisia and Turkey. Because 3 points will often be enough to advance, especially with a favorable goal differential (GD), and because a win is worth 3 points (and every team has one more game to play), it’s actually pretty hard to have been strictly eliminated at this point.3

This late-night newsletter is intended as a companion to our World Cup predictions page, and it might be helpful to open another browser window and go back and forth between them. For your convenience, though, here are our current projections for each team’s odds of finishing in each slot (1 through 4) in the group stage.

Read more

I spent a few days at the Winter Olympics in Milan this February. It was a relatively spontaneous decision; I had to be in Europe anyway. Attending wasn’t exactly cheap, but cheaper than we expected.1 We saw the men’s hockey semifinals2 — but one of the top sporting experiences of my life was watching Alysa Liu win the gold medal. I’m not a figure skating guy at all, but it was absolutely thrilling, and much better live than with the commentators snarkily talking over the performance on TV.3

The Women’s Free Skate is a long day, though — four hours and change from start to finish, basically 80 percent of which involves skaters who have literally zero statistical chance at winning.4 Nevertheless, I was struck by the poise of every competitor — most of them teenagers, out there completely alone — in what was undoubtedly a moment they’ll remember on their deathbeds. Being the best skater in, say, Estonia or Kazakhstan is an achievement unto itself, and the crowd reacted with great admiration, tossing flower bouquets and stuffed animals onto the ice.5

The World Cup is good, actually

Anyway, soccer’s World Cup is here in the United States. (And Canada and Mexico.) This plug is about as subtle as a FIFA “hydration break”, but quite a lot of you have subscribed to our World Cup forecast, and we really appreciate that…

I think it’s been an outstanding tournament so far, even as the narrative has shifted from a Day of Draws on Monday to a Day of Dominance on Tuesday. The tournament is averaging 3.1 goals per game so far, the most since the 1958 edition. And plenty of them have been from the world’s greatest players. If you’re even remotely a sports fan, how can you not be excited when Mbappe and Haaland both score a pair of goals — and then Messi tops them both with a hat trick?

There is, however, considerable debate over whether FIFA undermined the tournament by expanding from 32 to 48 teams.

Economically, this was obviously the right move. Despite some initial, basically reasonable skepticism about how FIFA was pricing tickets, the World Cup is just about the most popular product imaginable, and it only happens once every four years. Even games like Austria-Jordan are selling out, or coming very close to it. If products are flying off the shelf, you expand inventory. It would be completely irrational not to do that.

I’m not, by default, a maximalist. Our philosophy at Silver Bulletin is very much that more isn’t always better. And that goes for sports, also. I absolutely hate the NCAA tournament’s decision to expand to 76 teams, for instance. Look for a rant about that next March.

Nor does it increase the pool of potential winners. The teams ranked #33 through #48 in our initial World Cup projections — basically the beneficiaries of the expanded field — had a combined chance of about 0.2 percent of winning the tournament. Even the United States was by far a better bet than the entire bottom third of the field.

But do you not think it’s thrilling to Cape Verdeans or Haitians or Iraqis to be there to compete? Cape Verde actually drew with Spain on Monday. Iraq and Curacao lost, badly in the end, but at least their teams briefly leveled the score and gave their fans a moment of hope.

Does this come at a cost of competitive integrity? I’ve watched perhaps 75 percent of the World Cup so far, and I wouldn’t say so. Minnows are fun, and the matches have been feisty. The first weekend of the NCAA tournament is exciting precisely because of the possibility of seeing a 1-in-50 upset, even if UMBC or whomever was inevitably going to bow out in the next round.

Switching modes from fan to analyst, does the competitive state of international soccer warrant the expansion? In other words, is the talent gap narrowing between the Spains and the Cape Verdes of the world?

(I wouldn’t say the worst teams; there are 211 FIFA members, some of whom occasionally lose games by scores like 31-0.)

Not exactly. Our PELE ratings are (retrospectively) calculated all the way to the dawn of competitive international soccer in 1872, so we can look at the state of the sport as far back as the first World Cup in Uruguay in 1930. If you lined up the teams from #1 to #211 based on their PELE ratings at the start of each tournament, here’s what you’d get:

In 1930, there would hardly even have been enough FIFA members to fill out a 48-team tournament, and many of the prospective invitees declined to make the trip — by ocean liner. Since World War II, the gap has narrowed a bit but not that much. Some of the African teams are very good, as are Japan and South Korea in recent years. But the rest of the world (including the U.S.) hasn’t made consistent progress. Outside of Europe and South America, only the United States (in 1930), Korea (at home in 2002) and Morocco (in 2022) have ever reached the World Cup semifinals.

Still, though, why do you think college football teams frequently pay far-inferior rivals to travel to play them? It’s a feature, not a bug, for commercial sports that some teams have almost no chance. It can still be a win-win — the favorite usually gets a W, but the underdog pulls off an upset just often enough to keep things interesting. Meanwhile, FIFA and its broadcast partners have more advertising impressions to sell.

32 teams out of 48 advancing is too high a ratio

Few, if any, matches so far in the 2026 World Cup have failed to provide at least some entertaining moments.

The real problem, from a “game design” standpoint, is that FIFA’s new format has rendered the group stage more of an exhibition tournament. With two-thirds of the teams advancing, it almost takes more work not to advance than to do so. Spain, for instance, has more than a 95 percent chance of advancing in our model despite their unlikely draw against Cape Verde. There are really no more Groups of Death, since a majority of third-place finishers survive — the fourth-place team might be someone like Curacao, which lost 7-1 to Germany.

There are ways to keep the group-stage exciting with a 48-team format. FIFA could, in principle, have a 24-team knockout stage, with the 8 best finishers in the group stage getting byes. They aren’t going to do that, though, because England playing Uzbekistan in the Round of 32 rather than getting a free pass is still an opportunity to sell Coca-Cola products. (And England would probably lose on penalties anyway.)

A more underappreciated problem is that 12 groups flowing into 32 knockout-round slots creates both figurative and literal asymmetries. Two-thirds of the 12 first-place finishers are scheduled to face a grab bag of third-place teams based on what’s literally an indecipherable algorithm.6 The other four, though, are unlucky enough to play second-place teams in the R32, which means we could quite easily end up with Spain against Argentina (our two pre-tournament co-favorites) in the opening round should one team finish first in its group and the other in second. The old 16-team playoff format was much more elegant: every first-place team from the 8 groups faced a second-place one.

Go ahead and expand to 64

My preferred solution would probably be to see 8 groups of 5 or even 6 teams each (so, 40 or 48 teams total) with only the top two advancing. That would make the best group-stage matches extremely meaningful — advancing would be much harder — though it creates a risk of dead rubbers (i.e., games after both teams have been eliminated).

Probably the more natural solution, though, and the one that FIFA is more likely to adopt despite its occasional disavowal of the idea, is to simply double the previous 32–team format to 64. Instead of 4 teams x 8 groups (=32) you’d just have 4 x 16 (=64). The top two would advance from each group.7 You wouldn’t need any of these shenanigans involving the 3rd-place backdoor.

But wouldn’t this water down the group stage even further? I assumed so at first, but after running the numbers to calculate who else would have advanced this year under a 64-team format, I don’t actually think so.

Despite the 48-team format, a number of prominent teams are still missing from the World Cup, including countries like Nigeria, Italy, and Greece, which have large diaspora populations in the U.S., as well as sympathetic overachievers like Costa Rica and Cameroon.

Some of that is because FIFA has contradictory objectives. It undoubtedly isn’t happy that Italy missed the tournament (for the third time in a row!), but it also wants to grow the sport. If you went strictly by PELE ratings, then 26 of the 48 qualifiers would have been from UEFA (Europe) and 9 others from CONMEBOL (South America, excluding Suriname and Guyana). That doesn’t leave many positions for the rest of the globe, and many of the exceptions are either from North America or the oil-rich states of the Middle East, not really the developing world.

So what would a 64-team tournament look like? The conventional wisdom is that FIFA would punch up Europe and South America8, perhaps compensating for having gone in the other direction when going from 32 teams to 48.9 Still, you could keep things relatively proportional with an allocation like this:

I interrogated which teams would have been most likely to qualify this year if the tournament went from 48 teams to 64 using these allocations. It’s not an entirely straightforward counterfactual question because the whole qualifying format would have been different with more slots available. But I basically looked10 at which teams in each confederation came closest to qualifying this year but barely missed out. I even accounted for an expanded inter-confederation playoffs and randomized the results based on PELE odds. (Three of the four favorites advanced, but Ireland got unlucky and lost out to Honduras.11) No, I don’t care to debate the specifics of this, especially since this is meant as a hypothetical for demonstration purposes — though most of the choices are fairly straightforward, with some ambiguities arising from Europe, which had a weird qualifying structure this year.

The additional qualifiers were, in alphabetical order: Bolivia, Cameroon, Denmark, Honduras, Italy, Jamaica, Kosovo, Nigeria, Peru, Poland, Slovakia, Suriname, Ukraine, the United Arab Emirates, Venezuela and Wales.

Actually not such a bad group! Many of these teams would have been big draws in a North American World Cup and/or fun watches. If you look at PELE, the biggest gap is between the top 6 to 10 teams and the rest of the field. Team #32 isn’t that much better than team #48, and likewise, #48 isn’t that much better than #64, especially given the luck of the qualifying draw.

Having chosen the 64 teams, I used FIFA’s methods to randomly assign them to groups, abiding by a few requirements: 1) “Pots” were selected based on FIFA rankings as of November 202512; 2) No more than two UEFA teams or one team from any other confederation could go into each group; 3) Hosts (so the U.S., Canada and Mexico) were automatically put into Pot 113; 4) Inter-confederation playoff winners were automatically put into Pot 414 and 5) Since the assignment of group letters (i.e. A through L, now expanded to A through P) is basically arbitrary, I kept these the same for squads that were one of the 12 Pot 1 teams in the actual tournament15 (e.g. Argentina remains the anchor of Group J). Here are your new groups16:

Share

There are a few real stinkers here. The United States draws Wales, DR Congo and Jordan. We sure as hell ought to advance from that group. And France, instead of a relatively difficult Group I in the actual tournament, would face Panama, Egypt and Kosovo. But our new Group J is pretty good (despite containing three new qualifiers). Group H isn’t that easy either, with Spain, Turkey and Paraguay — and Spain would also get a rematch against Cape Verde.

But these groups aren’t obviously weaker than under the 48-team format this year. Moreover, with strictly two teams qualifying from each group instead of probably three, the consequences for missteps are much higher. If Spain had drawn with Cape Verde in our new Group H, for example, it would be in quite a bit of trouble with three decent teams in the group17 and only two passes to the knockout round.

You’re more than welcome not to like this expansion from 104 to 128 (!) matches, but the 2030 World Cup technically has six co-hosts18 and almost seems designed to facilitate it. So here's my bet: FIFA goes to 64, sells it as a one-off centennial gimmick (the first World Cup was in 1930), everyone complains, the complaints evaporate after another fun tournament — and then 64 is just how it's done.

Did that really happen? Although I might or might not have consumed a beer on a public walkway last night, my memories of the evening are fairly intact. And when I woke up this morning, the New York Knicks were still NBA Champions, making for the eighth distinct champion in eight seasons. Objectively speaking, I can’t claim to be surprised — I literally bet on the Knicks! But between the Knicks, Cubs and Red Sox in recent seasons, we’re starting to run out of sports curses — though apologies to our readers who are fans of a Canadian NHL team or the Buffalo Bills.

Silver Bulletin hasn’t been around that long, but if there’s anything resembling a tradition in these parts, it’s writing a piece of praise and analysis about the eventual NBA champion:

• 2023: The Nuggets were hiding in plain sight

• 2024: The Celtics are the prototype of a modern NBA champion

• 2025: Did the Thunder get too good, too fast?

This year’s edition is harder to write only because I’ve already exhausted a couple of the most obvious routes, having analyzed the Knicks’ transformation into a great basketball team before the Finals and written a personal tribute to the Knicks last year. (I “adopted” the team shortly after moving to New York City in 2009. I almost literally lived across the street from Madison Square Garden for more than a decade, so that’s my excuse.)

So I’ll instead give you five spare thoughts, to coincide with the number of games the Knicks needed to close out the series.

Subscribe now

1. Dynasty-or-bust NBA strategies are overrated

One constant in these championship write-ups is that analysis of what it takes to win the title can be myopic, with teams slotted into an overly narrow range of archetypes. Four of the past eight winners did not have a top-5 MVP finisher on their rosters — indeed, Jalen Brunson didn’t receive an MVP vote this season. The CW that a team can’t thrive in the post-season when a small guard is its best player also hasn’t held up well lately between Brunson and Steph Curry.

And recent postseasons haven’t been particularly friendly to favorites. The Knicks weren’t even that unlikely to win, having been +900 (implying a 1-in-10 chance) in futures markets to start the year; the Spurs (at +6600) would have been far more unlikely champions (indeed, the most unlikely winners ever). Nevertheless, among the eight non-repeat champions, only the 2024 Celtics were preseason favorites or co-favorites. This reflects a reversal of a streak in which the favorite won the majority of the time — 18 out of 34 tries — between 1985 and 2018.

This recent trend has contradicted the conventional wisdom that teams ought to basically adopt a variance-maximizing strategy of “dynasty or bust,” even being willing to tank entire seasons for the opportunity to acquire one or two superstars. We’ve even implicitly endorsed that attitude here at Silver Bulletin; our Future of the Franchise rankings are calibrated off our expectations for how many championships a team will win over its next 10 seasons, with everything else literally not mattering. (We had the Knicks 9th and the Spurs 2nd in the most recent edition.)

I’d like to point out, though, that even during that streak of favorite dominance, 18 out of 34 is not a particularly large sample size. What if the Timberwolves had drafted Steph Curry instead of Ricky Rubio and Jonny Flynn? What if the Bulls hadn’t traded for Scottie Pippen on draft day in 1987? The 1979 pick that brought the Lakers Magic Johnson originally belonged to the New Orleans Jazz. Maybe the league is changing, and modern salary cap rules are making dynasty building harder. But it seems plausible that favorites ran a little bit above expectations during this stretch.

Share

In poker, there’s sometimes a debate about whether it’s worth adopting lower-variance strategies to minimize your risk of ruin. For instance, a more conservative approach that theoretically yields $100 in expected value per hour but has a standard deviation of ±$500 might be preferred to a more aggressive strategy that nets $150 in EV but has a standard deviation of ±$1000. Generally speaking, though, these approaches are not advisable: the best way to avoid a losing streak that makes you go bust is to play as optimally as possible and to build up a cushion against the inevitable unlucky periods.

In the NBA, the situation is sort of reversed, with teams like the “Trust the Process” 76ers swinging for the fences and taking variance-maximizing approaches. But I wonder if that tactic is wrong, also, because it requires one too many things to go right. The Knicks have never been particularly rewarded by lottery luck; since taking Patrick Ewing #1 overall in 1985, the highest they’ve picked is 3rd (RJ Barrett in 2019). The Leon Rose-led Knicks, instead, could be typecast as a high-floor, lowish-ceiling team willing to “settle” for merely being pretty good, having indeed gone all-in on their current roster with the Mikal Bridges trade. Mitchell Robinson was the only important rotation player to be drafted by the team.

But sometimes “planning” to win 50-something games, rather than gambling on a dynasty, pays dividends too. For one thing, it makes the product more compelling for fans, as evidenced by the Knicks selling the most expensive tickets in NBA history during the Finals. Also, although “leaps” in ability are much rarer for veteran players than for young ones, they do happen occasionally. Karl-Anthony Towns, although he didn’t close out the Finals particularly well amid a mess of foul trouble, has transformed himself into a point-forward on offense and was basically as good as anyone in the playoffs at guarding Wemby. OG Anunoby was ranked as only the 40th best player in the NBA according to The Ringer in April, but by my estimation, he should be more in the 15-20 range.1 Sometimes coaching changes can help, too. It might be healthy for the NBA if more teams were willing to “settle” for being merely pretty good, and, under current cap rules, having a puncher’s chance of winning the Finals is better than most alternatives. Cases like the 2026 Knicks or almost the 2025 Pacers might become more common.

2. Momentum is underrated

A decade or two ago, stat nerds like me were fond of downplaying the importance and sometimes even the existence of momentum in sports, as illustrated by this XKCD cartoon:

But while it’s probably still true that most fans overrate the hot hand, I think the stat nerds were wrong to treat momentum as, at best, being a rounding error. And there have been two types of momentum on display in this NBA playoffs.

One was the Knicks’ incredible hot streak in the postseason — they went 15-1 after losing two out of the first three to the Atlanta Hawks, and finished with easily the best playoff point differential in NBA history.

While meaning no particular offense to, say, Dave Portnoy of Barstool Sports, it’s ironic that a certain class of pundit types are now downplaying the Knicks’ accomplishments.

Even if you weight 19 postseason games equally to 82 regular-season games, it really boosts the Knicks’ standing, taking them to +7.8 PPG across a 101-game season instead of +6.2 over 82. But that’s far too conservative an approach. The playoff sample came against strong opponents — no, the 76ers and Cavs aren’t world-beating teams, but they’re also not the Wizards or the Kings. And it’s recent data.

Basically, all the sports models I’ve built in recent years are Elo-type systems, meaning they continually revise a team’s rating after each game. How aggressively to update the rating is governed by a parameter called the “K-factor”, which can be determined empirically. In hockey or baseball — fairly random sports where even the best teams might only win two-thirds of the time — one game doesn’t tell you all that much. But in basketball, you should generally be pretty aggressive with your K-factors. One hundred possessions per team per game is a large sample, and the best teams can win in excess of 80 percent of their games.

Just look at some of the other teams on the list with the Knicks: they’re literally the consensus best teams in NBA history, including the Curry Warriors and the Michael Jordan Bulls. It’s very hard to sustain that sort of performance over a 19-game playoff sample without being really good. The Knicks weren’t even that hot, shooting 39.1 percent from three as compared to 36.7 in the regular season. It’s basically why I bet on them; I thought the market was treating them as a 53-win team and rounding up a bit, when there should have been a larger correction.

The other pertinent form of momentum is within-game momentum. I joked after the first quarter last night that the Knicks were taking a risk by only being down by 10. According to this Reddit post (I haven’t verified this independently), the Knicks are 8-62 in the playoffs since 2024 in games where they trailed by at least 14 points; the rest of the NBA is 10-110. You might have come across some data like this highlighting the Knicks’ incredible comebacks in the series:

I should say that I don’t quite believe these numbers as stated. Win probability models are a difficult problem, but as I covered in SBSQ #21, they probably understate the probability of big comebacks. Certainly live betting lines during these games didn’t have the Knicks quite so far out of it, and the betting consensus is usually better than the models.

There’s also the question of whether particular teams are more likely to make comebacks — or more likely to blow leads — based on experience or other factors. In other words, whether there’s a clutch factor. While I haven’t found much evidence of clutch performance in the NFL and only a little bit in baseball, the NBA is plausibly different.

One reason why is that teams score more often after a defensive stop in basketball — particularly on a fastbreak/transition possession. Although that’s a relatively subtle factor, it inherently creates some streakiness. A steal not only reduces your opponent’s expected points to 0 on their possession, it often also produces an easy bucket at the other end.

But also, NBA playoff basketball is at some sort of happy medium where it’s an incredibly high-intensity sport, but also one, I’d argue, where players do have time to think out there. It’s not a pure muscle-memory sport like baseball or hockey. It’s a sport where a player really can get into your head or the pressure of the moment can get to you.

My reporting for my book, as well as my experience playing poker in some occasionally extremely high-stakes moments, has changed my position on the clutch question a bit. I believe the best research suggests that human beings are essentially functioning on different operating systems in high-risk or high-stakes moments, that some people intrinsically adapt better to this than others, and that one plausibly gets better at it with experience — which brings us to this.

3. Playoff experience is underrated

As I mentioned in the Finals preview, the Knicks have played the most postseason games in the league since 2023, while the Spurs are one of the least experienced teams ever to make a deep postseason run. They’re basically getting a little bit of credit here for Harrison Barnes’s runs with the Warriors and not much else, and Barnes played less than 10 minutes per game during this year’s playoffs.

Joseph wrote about how impressed he was by the Spurs’ poise in their path to the Finals. I guess it’s one of those takes that “didn’t age well”, but if I’d assigned myself the Spurs’ half of the preview, I’d have written the same thing. As you may have noticed, SAS are also on the list for top postseason point differentials despite their loss to NYK, and their path was more difficult. (Yes, the opening round against the Blazers was something of a gimme, but the Wolves and the Thunder are extremely tough outs.)

“Poise” is not the term I’d use to describe the Spurs’ performance in the Finals, certainly, with a disastrous Wemby turnover in Game 2 and Wembanyama also sort of acting his age in the post-Game 5 press conference. But we’ve found in our past NBA modeling work that playoff experience can be predictive. It’s a difficult problem to study with the risk of various confounders3, but the 2026 Finals are going to be a high-leverage data point for anyone doing future NBA analytics work.

4. Jalen Fucking Brunson!

My taste in NBA players is pretty conventional. Off the top of my head, I’d list Steph Curry and Nikola Jokic as among my favorite players to watch, and Wemby was getting there before morphing into more of a villain role (at least from a Knicks’ fan perspective) in the playoffs.

And of course, Jalen Brunson.

There is one thing these players have in common. Curry, Wemby and Jokic are notoriously hard to compare to any other players in the history of the league. They’re the players who literally seem like a cheat code at times.

I’d argue that Brunson, who had 45 points in Game 5, fits that description as well, even though he isn’t quite in the same tier of stardom.

As a player, he’s a little bit contradictory. He has 10/10 intangibles, having now won a high school, NCAA and NBA championship, and he’s the son of a former NBA player. But he was drafted in the second round and underappreciated by the Mavericks. And he’s a very physical player despite not being particularly athletic by NBA standards. His ability to literally sort of force his way into decent shooting opportunities in the paint and in the midrange in some ways resembles a bigger player, but he’s rather generously listed at 6’2”. He’s plodding; literally no NBA player has the ball in his hands more often.

So who are Brunson’s best comps? Concerned that I’d missed someone obvious, I asked ChatGPT and Claude, but they came up with a list of players who don’t quite work. Isiah Thomas kind of fits, but Isiah had a much better pedigree as the #2 overall pick and was a significantly better defender and inferior shooter to Brunson. Chauncey Billups isn’t a bad suggestion, but is a tier below Brunson. Tony Parker? Also a relatively late draft pick, which helps, but a much worse perimeter player. The models even got creative, suggesting non-point guards like Adrian Dantley and Paul Pierce.

The point is that Brunson is an archetype. Some future scrappy, big-bootied point guard drafted late in the first round or early in the second after a deep run in the NCAA tournament will be compared to Brunson, not the other way around.

Advanced stats don’t particularly love Brunson, and he’s a negative on the defensive end by nearly everyone’s estimation. It’s hard to be a Top 5 NBA player unless you’re at least a decent two-way player. I do wonder, though, if there’s something about the variety of ways he can score (a little bit of everything except dunks; he’s had exactly 1 in-game dunk over the past two regular seasons and playoffs) that leads these models to underrate him.

If nothing else, Brunson is sort of inherently playing clutch basketball, given his tendency to burn through the shot clock. I don’t necessarily think this is advisable, and I prefer the Knicks when they play with more ball movement. But he hit 51.7 percent of his long 2’s this season (two-pointers of 16 feet or more). That provides a real tactical advantage on high-intensity possessions, of the sort that become more common late in playoff games. It sets a very high floor for avoiding wasted possessions. No NBA team maximizes all of its ~100 possessions, but between Brunson and their superior offensive rebounding, the Knicks come about as close as any team I’ve seen; maybe they’re going 96 or 97 for 100. Anyway, Brunson is probably the most popular New Yorker since at least Derek Jeter, and deservedly so.

5. New York Fucking City!

I’ve decided not to make a tribute to New York City the principal theme of this newsletter. That just felt a little bit too clichéd, even though, in a very tangible sense, I’ve invested more in NYC than anything save my relationship (20+ years) and my business. But the cliches are fucking true. There’s really nothing that unites New York like the Knicks, except maybe everyone going out to run/drink/mingle on warm early-spring/late-summer days, and lately we’ve had a lot of both. And there’s no better place to live than New York when the city is at its best. Here was a representative scene in Fort Greene last night:

While the Yankees and Giants have won championships since I moved here in 2009 — and the Rangers, a bigger deal locally than you might assume, have come close — there hasn’t really been any moment like this except maybe the 2020 election (when everyone sort of came out of their COVID shells). In baseball, there’s too much of a Mets contingent, and rooting for the Yankees does feel like frontrunning. The Giants play in New Jersey, and hockey is just a little bit too obscure. The throngs of multi-ethnic, multi-everything crowds are only going to emerge for truly monumental moments, like the Knicks winning their first championship in 53 years.

One thing you might not get about NYC is that it’s not a particularly gate-keepey place. We wait on line because it’s crowded, but it’s not a city of velvet ropes (or at least not if you’re “doing NYC right”). Some 37 percent of the population is foreign-born, and more than half was born outside New York state. Although I do have some NYC heritage4, I never felt like I wasn’t a “real New Yorker” even when I had less experience under my belt in the city. It’s an extremely self-selected city, and if you select into the high cost of living and frequent near-chaos, you’re allowed to select into the Knicks. That’s part of the deal, and given the Knicks’ legacy, it’s not as though this has been a particularly high ROI choice over the years.

I’ve been getting ribbed by my friends for attending Game 3 — The Trump Game — because it was the only game out of their past 16 that the Knicks lost. It wasn’t a cheap ticket, though I got a pretty good value relative to prevailing market prices. I suppose I felt like New York had something to prove and I wanted to be there for it. Another thing that people may not get is that New York is a loose culture, not a tight one; we’re not necessarily that big on officially following the rules, from jumping subway turnstiles to tolerating illegal weed stores. But it’s also one of the most efficient and productive places on earth. So I wanted the narrative to be that, for all the chaos surrounding the event, universally regarded as a “shitshow” by the fans waiting on line with me (about an hour and fifteen minutes from lining up on 6th Ave to getting into the MSG seating bowl) — that New York would get it right in the end.

The crowd at Game 3 wasn’t particularly great. It seemed to lose momentum by the second half, perhaps because of the choppy play dictated by the officiating and the long wait to get in. Sitting toward the front of the 200 Level, I basically had a big responsibility for whether the fans behind me were sitting or standing. At one point late in the fourth quarter — this is very much not in character for me5 — I turned around and implored people to stand up, screaming something like “this is the fucking NBA Finals!”. “We” didn’t get the outcome we wanted that night, and the walk home through corrals of police barricades was sort of a downer. But waiting five more nights was a small price to pay — let’s not talk about the Ticketmaster fees.

This newsletter aims to cover a lot of bases. We’re aware there’s a lot of political news and sports news. We’re working overtime to cover these beats; I’m not sure we’ve ever published so much in such a compressed time period before.

And life can be wonderful in early June, especially if you live in a temperate city like New York. So we’re also aware that even the substantial number of sports fans on the Silver Bulletin mailing list are not necessarily grappling with the fact that the World Cup will begin on Thursday. The opening games are in Mexico; the United States’s first match is set for Friday night in Los Angeles. (See here for a more detailed outlook on the U.S.’s prospects.)

However, I did want to give you one more pitch for our World Cup predictions. We’ll update these projections every day throughout the tournament. But today’s newsletter provides some bonus World Cup content inspired by reader suggestions. I’ll take advantage of our model to answer two common questions:

Subscribe now

1. What is each country’s most likely path to the World Cup title?

2. And if you’re thinking of watching or even attending a match, which group-stage games will have the greatest impact on the eventual outcome of the tournament? And might be most worth buying tickets for?

Paths to the World Cup title

With a 32-team knockout-stage bracket in place for the first time ever, there’s basically no avoiding that your favorite soccer team will eventually have to defeat some tough opponents. To be honest, the group stage is rather forgiving, with two-thirds of teams set to advance into the “playoffs”. However, our simulations suggest that group-stage slots are far from equal. There actually is an advantage in most cases to winning your group. And although the top 8 of 12 third-place finishers will also advance — this World Cup format is practically giving out participation trophies — these paths are generally pernicious, yielding tougher R32 and R16 opponents.

So here’s one fun way to look at the group stage. Conditional upon finishing first or second in the group, or qualifying from the 3rd-place slot1, what are your team’s most likely opponents in every subsequent round? And what are its odds of taking home the FIFA World Cup Trophy?

This is one of those times when I’m going to show you the data first and explain it later. As a warning in case you’re a fan of Uzbekistan or something, I’ve restricted the outlook to the top 30 among the 48 teams. This basically covers everyone, including co-hosts Mexico, Canada and the United States. But the bottom 18 teams are so unlikely to win that the simulation results are mostly statistical noise, even though we ran 200K simulations today instead of our usual 100K.

Read more

It’s been a satisfying few months as a sports fan. “My” Knicks have their best chance at a championship since 1973. And Jack Hughes’s overtime winner over Canada in the Gold Medal Match in Milan represented a triumph for the men’s hockey team.

So is it too much to ask that this is also the year when the U.S. men’s soccer team breaks through, with the World Cup set to kick off this week on North American soil?1

The 23 previous hosts2 have a pretty good track record, with 6 outright wins (though none since France in 1998). Another 7 hosts made at least the semifinals. So an optimistic spin on this data might suggest that the U.S. has a 50/50 chance of making the final four, something we haven’t done since the inaugural World Cup in 1930.

But that’s not what the smart money thinks. Prediction markets give the U.S. only a 10 percent chance of reaching the semifinal. No spoilers, but our forecast is in the same broad range.

Most host nations, of course, have better soccer legacies than the United States. (Or at least, better legacies in men’s soccer; our women’s team has won 4 of the 9 Women’s World Cups.) The two tune-up matches the U.S. played in advance of the tournament are typical of our not-quite-arrival on the scene. We beat Senegal 2-1 last week but then lost by the same scoreline to Germany in Chicago yesterday, conceding a goal in the second minute. These aren’t bad results: Germany is our 6th-ranked team in PELE and Senegal is 16th. But it’s more of the win-some, lose-some outcomes that have long characterized American soccer.

A future that has never arrived

Soccer has long been considered the “sport of the future” in the United States, but that’s sort of a running joke: the future has never really arrived. Soccer exploded as a youth participation sport in the 1970s and 1980s, and by 1978, the New York Cosmos were drawing almost 48,000 fans per game at Giants Stadium. But by 1985, the North American Soccer League was defunct.

Perhaps there was a little spark of something there, because American soccer had truly been through its Dark Ages. The U.S. failed to qualify for a single World Cup between Brazil 1950 — when we miraculously beat England 1-0 but lost our other two games — and Italia 1990, when our return to the global stage was marked by a 5-1 loss to Czechoslovakia.

What might be less well known is that American soccer looked more promising in the early years of the sport.

Silver Bulletin’s PELE ratings are calculated back to the first international match in 1872. And as you can see above, they began with a relatively optimistic prior about the United States, with an initial PELE rating of 1904. That would rank us about 15th in the world today, in the same general vicinity as Turkey, Uruguay, Belgium and Italy.

That initial rating is based on three factors: the size of our economy (always enormous), the year that we first joined FIFA (1914, nine years before Brazil did) and a regional coefficient. The regional coefficient does a lot of work, and there’s room to debate whether it’s too high or too low. But the USMNT lived up to the model’s expectations in the early years. Our first two World Cup matches in 1930 were 3-0 wins, producing an all-time high rating of 1958 on July 17, 1930, before we got crushed 6-1 by Argentina in the semifinal.

However, the U.S. didn’t play any international matches between 1937 and 1947. Early attempts at domestic leagues never really cohered. And with a World War going on, we had more important things to do. But when our team returned after the war, it was in much worse shape. We lost our first six matches after WWII by a combined scoreline of 36-2, including indignities such as a 11-0 loss to Norway. Our 1-0 victory over England in 1950 is one of the World Cup’s greatest all-time upsets. But it was an upset precisely because we weren’t very good. Indeed, that’s the only World Cup match we won between 1930 and 1994.

Football and an isolationist streak slowed soccer progress

So what went wrong? You could chalk it up to soccer being viewed as a “foreign sport” after World War II, but the game was invented by England, our ally, not the Axis powers.

Two other factors probably played a larger role. One of them is football — or excuse me, American football. The gridiron codes developed in the late 19th century at places like Harvard and Yale steered our version of the sport in a different, more rugby-like direction. While the optimal body types for American football and soccer aren’t really the same3, football has long been the focal point for amateur athletics.

And the NFL developed into a huge commercial enterprise in the 1950s and 1960s during the Dark Ages for American soccer. Not to mention other homegrown sports like baseball and basketball. The United States is not exactly lacking in athletic prowess, as our women’s team and our success in other sports show.

But it’s fair to say that soccer has long had a reputation in the U.S. as an “immigrant game”, with our strengths in regional and ethnic pockets of the population. In theory, that might nevertheless play to our strengths. Even amid a recent period of immigration backlash, the U.S. has more than 50 million foreign-born residents, triple that of any other country. It’s hard not to notice, though, that our soccer success has tended to rise and fall with our immigration levels, the Dark Ages coinciding with a more isolationist streak before and after World War II.

The U.S. finally qualified for the World Cup again in 1990, though we lost all three group stage matches. In 1994, we got an automatic berth as hosts and turned in a more promising performance, beating Colombia in the group stage, albeit on an infamous and tragic own goal, before bowing out respectably 1-0 to eventual champion Brazil in the Round of 16.

The next decade or so represented steady upward progress. Yes, the 1998 team was overmatched in France. But in 2002, we won our first and heretofore only World Cup knockout stage game in Jeonju, Korea, defeating rivals Mexico dos a cero. Meanwhile, MLS was established in 1996 and was proving more stable than past leagues like the NASL.

Since then, however, we’ve been treading water, with highlights like Landon Donovan’s dramatic late goal against Algeria to advance us to the knockout stage in 2010 and lowlights like failing to qualify for the 2018 World Cup at all after washing out against Trinidad.

This year’s team isn’t on a particularly promising trajectory. We hosted Copa America in 2024, which had typically been held in South America, but failed to advance out of a relatively weak group featuring Bolivia, Panama and Uruguay.

Meanwhile, we haven’t beaten any team in the current PELE top 10 since a 4-3 win over the Netherlands in 2015. So we can beat the Senegals and Algerias of the world, but not the Germanys or Brazils — although we do seem to have a knack for drawing England.

Is MLS part of the problem?

If you’d asked me in 2002, I’d have expected MLS to become a much bigger deal by 2026. And the league is successful in some ways. MLS now has 30 teams, and it’s a solid draw, attracting more than 20,000 spectators per game.

However, the league lacks traction as a national focal point that aspiring athletes dream of playing in. Google searches for MLS-related terms are only one-fifth as high as the NHL — and less than 1/30th the volume for the NFL.

MLS has an unusual, centrally-owned structure, with the 30 franchises being more like true franchises in the McDonald’s sense, with strict spending limits and a risk-averse approach. Exceptions famously can and have been made for aging stars like Messi and David Beckham. Overall, though, the combined player market values for MLS is “just” €1.4 billion, according to Transfermarkt. For comparison, PSG (€1.37b), Man City (€1.32b) and Arsenal (€1.25b) each have nearly as much player value on their rosters by themselves.

It’s surprising, given the extremely lucrative market for sports franchises among multi-billionaires, that in a famously capitalistic economy, the United States doesn’t have a few teams that are attempting to compete with the world’s best. LAFC and Inter Miami have franchise valuations of $1.2 billion — that’s the estimated market price for franchise resale, not the Transfermarkt player valuations described above — but that’s lower than even the NHL’s Winnipeg Jets. A rival to MLS called USL Premier, which would feature full relegation and promotion, has been proposed, but is in the very early stages.

But aren’t our players increasingly successful in Europe? Although Christian Pulisic has had an arguably underachieving career, the answer is basically yes. In addition to factors like region and GDP, PELE uses Transfermarkt valuations for a country’s top 23 players to help set its priors. Even after adjusting for persistent inflation in player values, the value of the U.S. roster has roughly tripled since 2005. It’s also become more skewed toward attacking players, in contrast to our history of being stronger at defense and goalkeeping.

However, the twin development pipelines of MLS and Europe don’t necessarily lead to a highly coherent approach to international matches. Our 2026 World Cup squad is split between 8 MLS players, one in Liga MX, and 17 on various European teams. No two members of the 26-man roster play on the same club team.

How much this matters is hard to know. In our research for PELE, we found that rosters heavy with players from domestic club leagues slightly overperform, but the effect is minor. The bearish case for this model is Africa, which has lots of soccer talent, but almost all of which plays in Europe. The only African country ever to reach a World Cup semifinal was Morocco in 2022. However, Brazil and Argentina export most of their best talent to Europe and have been perfectly fine, obviously.

Why FIFA overrates Team USA

These mixed signals make the United States one of the harder teams to rate. The World Football Elo Ratings have the U.S. 38th. PELE has them 29th. But the FIFA rankings, which are also based on an Elo-type system, have them 17th.

What accounts for the difference? The comparatively more optimistic rating for the U.S. in PELE than in the World Football Elo Ratings reflects the effect of the priors that I described. Basically, PELE thinks the U.S. should be better than we’ve seen from match results alone. This prior has some predictive power, especially given that high-stakes international matches are relatively infrequent.

The aggregate Transfermarkt value of our optimal 23-man roster ranks 17th in the world; that’s the main benchmark PELE is using. We also calculate a “GDP prior” based on GDP, region and soccer legacy (as measured by the first year in FIFA4), and which plays a secondary role in the system.5 There we rank even higher: 9th. Granted this is tricky, because the regional variable carries a lot of the load, and there aren’t a lot of good comparison points for the United States. 6

So basically, PELE endorses the notion that the USMNT “should” be better at soccer than it is. Probably not a top-flight team: the soccer hegemony of Latin America and Europe has been quite persistent over time. But PELE is always trying to nudge the U.S. a little upward, and we are often failing to live up to those expectations.

What about our #17 ranking in FIFA? Well, it’s wrong. That’s not to say we aren’t capable of playing that well, and 17 versus 29 isn’t that big a gap. But FIFA overrates the quality of our resume from match results alone.

It’s wrong for a simple reason: FIFA doesn’t account for home-field advantage. Since 2015, the U.S. national team has played 134 of its 179 matches on home turf. That’s 75 percent, roughly double the average of 37 percent for other participants in the 2026 World Cup.

We host a lot of friendlies and regional competitions like Copa America. These matches are often well-attended and popular. But home-field advantage makes the going a lot easier in soccer. In the World Football Elo Ratings, a home match is worth 100 Elo points. Ignoring draws, that increases the win probability between two evenly-matched teams to 64 percent for the home team. PELE uses a considerably more complicated system for home field, incorporating travel distance, altitude, and long-term performance for home teams both collectively and individually, so each team gets a customized factor. But the United States’ home rating (+83) is fairly high.

Home-field is our biggest asset until proven otherwise

However, we’ll get to take advantage of home-field advantage in the 2026 World Cup. All of our group-stage matches will be played at home, and the knockout-stage draw has also been set up in a favorable way for us. Should we make the Round of 32, it’s guaranteed to be a home match; and then the U.S. hosts everything from the quarterfinals onward.

Put another way, that #17 FIFA rating inflates our stature because it’s largely based on home matches. But since essentially all of our 2026 World Cup matches will also be played at home, it’s a pretty decent approximation for how tough of an out we’ll be in this tournament. Indeed, we’re the 16th most likely team to win the World Cup according to PELE. Although those chances are only around 1 percent, they’re considerably higher than they’d be in a tournament played in Qatar or Russia or Spain.

That leaves us in a position more like that of the 1980 Olympic hockey team — not this year’s Hughes-brothers-led edition.7 Winning our group will be the first step, since that would set us up to play a 3rd-place team in the Round of 32. Our most likely R32 opponents are Bosnia, Egypt and — this one would be interesting — Iran. Those are winnable matches. Beyond that, we’d probably need a Miracle on Grass. But stranger things have happened in soccer.

Share

It’s a busy week here; thanks as always for reading Silver Bulletin! While we’re not neglecting politics, we’re very excited about this year’s matchup in the NBA Finals, which start tomorrow. I made my glass-half-full case for the Knicks on Sunday; today, it’s Joseph’s turn to make the argument for the Spurs.

Joseph and I also chatted with Nate Duncan of the Dunc’d On podcast yesterday — one of our absolute favorite shows — about the Finals and other pressing NBA storylines. So you’ll see the video first (I believe this is Joseph’s video debut at Silver Bulletin!) and then his Spurs story.

One other heads-up: with rosters being finalized today, our World Cup forecast will launch imminently, we believe tomorrow, but don’t kill us if it slips to Thursday. When we have a short-term feature of this nature, we increase pricing for newly initiated monthly subscriptions. This does not affect annual pricing or people who are already subscribed. We just wanted to disclose that ahead of time; the pricing won’t change until the World Cup forecast is published tomorrow or Thursday. —Nate Silver

Subscribe now

Our NBA chat with Nate Duncan

The case for the Spurs

by Joseph George

If there was any doubt, Saturday night’s Game 7 made it obvious that this year’s Western Conference Finals was one of those series — the stakes were extraordinarily high, a number of the league’s main characters left their marks, and no team ever felt like it established true control.

The Spurs and Thunder had been on a crash course for a long time. Oklahoma City has occupied the role of “best positioned team in the NBA” since 2022 at least, but the Spurs’ lottery luck vaulted them into the same tier of short-term and long-term contention. Both teams sit in roughly the same age window, ranked #1 and #2 in record and simple rating system, and have consistently held the top two spots in our Future of the Franchise rankings. As of a few months ago, there honestly wasn’t much deliberation at the top end of FotF; the Spurs were clearly #2 ahead of the rest of the league, but the Thunder were clearly #1. This series obviously calls that into question.

There’s not a lot of love lost between the two franchises either. The matchup doesn’t carry the pedigree of a Lakers-Celtics, but both teams have loyal, dedicated hometown fanbases and a real claim to being the model for a modern basketball franchise. Those past matchups were notable for the depth of talent they featured and the narratives around those teams. The Thunder were young, loud, and stupidly athletic; the Spurs were the dynasty in its long, graceful decline, buying time with system and pedigree.

The newest iteration of the Thunder-Spurs rivalry has only had the first of what should be many future playoff series, so while we don’t know exactly how this story will go, the historical parallels are easy to see. The emergence of young players during the series, like Dylan Harper and Jared McCain, forced both teams to make heavy adjustments. The Thunder aren’t much older, but they played the part of the battle-tested veterans. The Spurs, whose three best players in this series were in their third, second, and first seasons, didn’t blink, answering every punch with a swift counter.

How did we get here?

That lack of experience was certainly one of the bigger storylines during this series, and it’s even more dramatic than I had previously assumed. The Spurs aren’t just young and inexperienced for a Finals team: by most measures, they’re one of the most inexperienced teams to ever make the playoffs. For reference, the closest historical analog is the 2012-13 Warriors, who fell in the second round to the Tim Duncan-led Spurs.

There is the question of how they’re overcoming this disadvantage. Empirically, playoff experience does matter, but perhaps its effects can be more dramatic for certain types of teams. The Spurs have been particularly resistant to some of the typical issues that beset younger teams in the playoffs, like suffering from long scoring droughts or struggling to make consistent rotations. Instead, their overall play has actually improved — over a sample of 18 games in the playoffs, they’ve maintained a net rating of +11.0, a decent improvement over their +8.4 mark in the regular season.

Now, I do recognize that I’m several paragraphs into an article about the Spurs and I haven’t mentioned Victor Wembanyama. Part of that is the point — this run has been a team achievement in a way that's easy to miss when one of the players is Wemby. But it's hard to overstate just how unprecedented he is, both in his on-court playstyle and his accolades at such a young age. Wemby is, to put it plainly, complete madness to watch. It’s not just the step-through dunks from inside the free-throw line, or the pull-up threes over other seven-footers. It’s the poise he shows in anchoring team defense, and the growth he’s shown in setting up a very organized offense this season — both genuinely rare for someone at his developmental stage.

When I wrote my debut byline about Wembanyama back in October, I postured about whether he could make the MVP leap, which felt like a longshot in the moment. While he didn’t take home the award this year, he has certainly placed himself in the “best player in the world” argument, if not already grabbed the title outright. Wemby belongs to a rare class of rim protectors and can outright take over the offensive end at times.

Somehow, Wemby’s defense has gone up a notch in the playoffs — he has the best rim deterrence on/off of any player ever with at least 500 minutes in the playoffs — and the Thunder had to resort to some interesting strategies to keep him away from their drivers.

Wemby hasn't had to carry it alone, though. The other Spurs went 12-6 in games without him this year and have also stepped up big time. I was worried Stephon Castle’s looseness with the ball might be too much to overcome after Game 2, but he didn’t let it hamper his aggression, which proved to be the correct gamble as the series played out. It will probably be a bigger deal in future seasons, but there’s something to be said about how he was able to find other ways to impact the game as the series carried along.

The Spurs’ X Factor is likely their youngest player, Dylan Harper, who looks like he’ll be one of the best guards in the league very soon. He’s already creating at historic levels near the rim. Some of the greatest point guards of all-time have had similar (indeed, often slightly worse) profiles to Harper at the same age, and most of them never had his high defensive floor. Harper was already one of the highest volume self-creators at the rim this regular season, with the only hole being his inability to draw fouls consistently. In the playoffs, his free throw attempts per 100 possessions nearly doubled, which is incredibly rare for a rookie guard. (They don’t typically get a generous whistle.) If this continues to progress, he could return All-NBA level value while still on his rookie deal.

Harper’s ability to push in transition and in the half-court gives San Antonio an additional release valve, and it showed against Oklahoma City — in multiple games this series, the Spurs went to Harper as a creator at the end of the shot clock. At the end of Game 7, Harper launched an absolute bomb over SGA, which quieted the Thunder crowd.

It reminded me a little of a young James Harden in 2012, before most had pegged him as a guy who would command the attention of an entire defense. I don’t think Harper is there just yet, but I keep coming back to the fact that the Spurs are getting this from a rookie who isn’t even their best young player. If there’s a team that could show a lack of experience doesn’t mean a lack of poise, it’s this one.

Do the Spurs have an answer for point-KAT?

Plenty of people seem to think the Western Conference Finals already settled who’s winning the title. I don’t exactly buy it, and underrating the Knicks feels backwards. We almost never see playoff runs this dominant. New York chopped up a decently tough schedule, and it’s a signature of that dominance that their opponents look like roadkill in the rearview mirror.

As Nate wrote on Sunday, the Knicks have KAT-ified their offense since their early struggles against Atlanta. I’ve actually long wondered whether KAT’s ideal role is exactly this — less a usage-sink secondary scorer, more an oversized creator running the offense through his passing, like an oversized attacking midfielder in soccer, a Ruud Gullit of sorts.1

Towns is driving the ball at similar rates to the regular season, but it’s clear that he’s found a better cadence and angles in the playoffs: his finishing percentage on drives and his passing percentage have gone up 6 percent. This isn’t the first time wingifying a big man has had good results — in 2023-24, Joel Embiid transitioned more from baseline post-ups to face-up elbow actions, which made him one of the best passing bigs in the NBA almost immediately. It’s possible that KAT’s involvement in the offense unlocks a dimension for the Knicks that the Spurs aren’t entirely prepared for.

As a result of KAT’s self-creation jump (and his floor spacing), Wemby will likely be assigned to Josh Hart to avoid pulling him out of the paint too often, which puts pressure on the Spurs’ forward depth. Wemby’s rim presence helps, but Carter Bryant, Devin Vassell, and even Castle will have to spend sizable stretches on Towns, and avoiding foul trouble becomes a very tangible goal if they want to keep KAT from leaving a mark on this series.

Still, this isn’t a complete positive for the Knicks. KAT has often struggled with body control, and the threat of Wemby roaming will make it hard for him to find the same angles to the basket he’s enjoyed over the last two series. It’s possible the series swings the Spurs’ way as a result of Towns getting into his own foul trouble. We can sort of reliably expect KAT will have at least one of those games — he was the worst big man in the NBA this year when it came to offensive fouls, and a matchup that forces him to put the ball on the deck against smaller players is exactly where he gets into trouble, picking up cheap charges and player-control fouls that send him to the bench. If the Knicks can’t play Towns as much in the second or third quarter, the Spurs will try to bury them in defensive stops.

The variability of these matchups makes it hard to know what version of the Knicks the Spurs will have to face. During the regular season, the Knicks had a tendency to break down offensively at times, usually as a result of poor communication or bad process. The playoffs have featured less of that, but they also involved back-to-back rounds of pretty weak guard defenders (Donovan Mitchell, Tyrese Maxey, James Harden). Brunson may have a harder time trying to create against the Spurs guards and wings, as they contain a bit more length.

On the offensive end, the Spurs will have to do with less Wemby self-creation away from the basket. If I’m correct, the Knicks will have OG Anunoby on Wemby, in hopes that OG’s relatively low center of gravity will make it hard for him to create much on post-ups away from the basket. Luckily for the Spurs’ guards, there are real weak spots for them to attack in the Knicks’ rotation. If they can force Brunson on Harper or Castle, they’ll be in cruise control. Of course, the Knicks’ wing stoppers will make it hard to get that match up very consistently, but on the margins this feels like an area to exploit.

I’d expect that the Spurs will try to take advantage of lobs for Wemby, and their transition offense will be in full effect against Brunson and Towns, but I’m still not totally convinced that this is a great series for either team. The Spurs seem like a tougher matchup for New York than Oklahoma City (with their injuries) would have been, but there are real levers that New York can pull to stretch San Antonio thin.

If I had to estimate, this series goes six or seven, and I’d lean San Antonio. Obviously, a lot will be made of the Spurs’ inexperience if they lose, but I’d push back on that framing in advance. The Knicks don’t need San Antonio to play down to its age to win this series; these are two genuinely good teams, and in their current form New York might simply be the better one. For the Spurs, the youth that looked like a liability back in May has been the whole story of this run, and I don’t see the Garden being the thing that finally cracks them. Still, the mechanisms are there for New York to turn this into a war of attrition, and I wouldn’t be surprised if they walk away with the trophy.

With San Antonio finishing off Oklahoma City, the Spurs and New York Knicks are set to face one another in a rematch of the 1999 NBA Finals that could be the most fun matchup in years. This is the first of our two-part NBA Finals preview; I’m making the case for the Knicks, while Joseph George took on the Spurs in Part II. We also did an NBA Finals Substack Live with special guest and friend of the newsletter, Nate Duncan; you can find that in Joseph’s story.

Subscribe now

I can’t claim to be a long-suffering Knicks fan, but I’m not exactly a bandwagon-jumper, either. I moved to what is technically northeast Chelsea in 20131, but which is really more of a no-man’s land between Midtown and various places south2, a micro-non-neighborhood defined by its abundance of parking garages, “Irish” sports bars, illegal weed stores, and most of all, its proximity to Penn Station and Madison Square Garden. When a team’s stadium is almost literally across the block — for a time, my apartment was constantly bathed in red light from an extremely vibrant Rao’s Pasta Sauce ad on the MSG outdoor billboard — and that team is notoriously, chronically underachieving, you are provided with the option to adopt it in midlife, as clearly articulated in §14.2.7 of the 1955 Sports Fandom Accords.

Linsanity also helped whet my appetite. Those days in February 2012 were proof of concept for how the city would go absolutely crazy for the Knicks. They also felt like something out of the movie Awakenings: glorious, but glorious precisely because they were obviously too good to be sustainable. Indeed, Lin played just one season with a team, with the Knicks declining to match an offseason offer sheet from Houston. Fans might have been ready to get over it following a fluky 54-28 record in Anthony’s third season with the team in 2012-13. But after a second-round exit at the hands of the Pacers, the Knicks never again made the playoffs for the rest of Carmelo’s tenure in New York.

The Jalen Brunson Era has obviously been a revelation as compared to the false dawns that preceded it. (I haven’t even mentioned Kristaps Porzingis.) But my take on the current iteration of the Knicks has long been that they were a canonical, high-floor, limited-ceiling team: a good watch but bound in by middling defense and roster inflexibility after a series of all-in moves. I supported the replacement of Tom Thibodeau with Mike Brown largely on the grounds that the Knicks didn’t have many other options to improve.

I actually did bet on the Knicks to win the Eastern Conference before the playoffs began and doubled down while they were in the midst of their 140-89 shellacking of the Atlanta Hawks in Game 6 of the opening round. (The odds were good, though my proceeds are quickly being squandered on World Series of Poker buy-ins). But I assumed that would be as far as it went before a 4-1 shellacking against the Thunder or Spurs.

I’ve changed my view. I think the Knicks could actually win the NBA title — and win it fair and square. Not just because of incredibly hot shooting — though that would help — or a Spurs injury, in other words. I wouldn’t bet on them at even money, especially with Game 7 scheduled for San Antonio’s Frost Bank Center, not MSG. But I think this might be a great team, not merely a very good one.

What to make of a truly insane playoff run

You’ve probably seen stats like this: the 2025-26 New York Knicks currently have the best playoff scoring margin in NBA history at +19.4 (!!!!!!!) points per game. It’s been such a dominant stretch that all three of the Knicks’ vanquished Eastern Conference opponents have to deeply reconsider their futures — the 76ers fired Daryl Morey, indeed.3 Of course, the playoffs aren’t over yet, and I wouldn’t count on the Knicks beating the Spurs by 19 points in the Finals. They’re in truly elite company, though: Jordan’s Bulls, Kareem’s Bucks, the Shaq-Kobe Lakers, and Steph’s Warriors are the others in this stratosphere. Wemby’s Spurs have also outscored their opponents by double digits so far in the playoffs, though, I’d note.

But a hot streak in basketball isn’t merely a curiosity; it can also be predictive. Here’s some additional context that I’m pretty qualified to provide, having spent thousands of hours developing sports rating systems like RAPTOR, COOPER and PELE. Recent performance should be weighted fairly heavily, especially in the NBA, which is perhaps the most deterministic sport since each team gets around 100 possessions per game. In the NBA, 14 games is not a small sample and especially not in the playoffs. If you’re calibrating your expectations for the Knicks based on their 53-win regular season, you’re doing it wrong.

What’s unusual, though, is that the Knicks aren’t in either category of teams that usually make this sort of leap. Middle-aged in basketball terms, the Knicks are not a young team that is suddenly coming into its own. But they’re also not a team that has so proven its bona fides that they can lollygag their way through the regular season in the style of the 2000-01 Lakers. In fact, that’s sort of the antithesis of Brunson’s approach, or Thibodeau’s or Brown’s.

The Knicks do have a lot of playoff experience, however, something we’ve found in the past is also predictive. In Brunson’s four years with the team, they’ve played 56 playoff games; only the Celtics have played more during this period, and the Knicks will surpass them when they take the court for Game 2. They have a 35-21 record despite catching what you could argue to be various bad breaks. They did also win the NBA Cup this season, defeating (drumroll) the Spurs.

The Knicks’ defense suddenly got way better

The Knicks’ 53-29 regular-season record was similar to last year’s 51-31 and 50-32 from two seasons ago. But that, plus a midseason slump when they went 2-9 over 11 games in January, tends to obscure some real improvement. The difference came on defense: the Knicks’ regular-season defensive rating jumped from 14th in the league last year to 7th this year. Moreover, the defense got better as the year went on and Brown had more time to shape the approach.4 The Knicks allowed 114.9 points per game in games through Jan. 19, the last day of that slump. But after limiting the Brooklyn Nets to just 66 points on Jan. 21, they only allowed 104.9 PPG for the rest of the regular season — and it’s been 100.6 so far in the playoffs, with the Knicks having the best defensive rating in the league in the postseason.

“Very good offense, average defense” is a plausible formula to win the Eastern Conference. “Great offense, good defense” is what they’ll probably need to beat San Antonio.

The fiercer defense also makes the Knicks feel much more Knicks-y. The Knicks have always risen and fallen with their defense; the Patrick Ewing teams actually had a below-average offense even in the years they reached the Finals, but a defense that would beat you into submission in a manner that perhaps only the Bad Boy Pistons ever matched. (The average scoreline of the 1999 NBA Finals, also against San Antonio, was Spurs 85-Knicks 80.) The strengths of the 1970s championship teams were also concentrated on the defensive end.

But is the newfound elite defensive performance sustainable? Listen to enough NBA podcasts, and you’ll frequently hear the take that the Knicks have two massive defensive holes to patch: Brunson at the point and Karl-Anthony Towns in the frontcourt. So even with a trio of very good “wing-stopper” defenders in OG Anunoby, Mikal Bridges and Josh Hart, they’re sort of treading water.

The critique of Brunson is fair. He’s not a black hole like Trae Young out there, and he gets his share of steals, but his low center of gravity does not really translate on defense according to either the eye test or the advanced stats. However, I’d argue that the Knicks have one big defensive liability, not two. According to EPM, Towns had the best defensive season of his career and was a slight defensive positive overall.

Some of that is because KAT is a terrific rebounder, averaging 13.8 rebounds per 36 minutes this year, a career best. Rebounding has become underrated among analytical types: it doesn’t count as a defensive stop until you grab the board.

Admittedly, KAT is perhaps the weirdest player in the NBA, spectacularly talented without being particularly athletic or confidence-inspiring. He is almost never described as a “unicorn,” despite probably being the best-shooting 7-footer in NBA history. Towns had a reputation coming out of college for being a “very good shot blocker”, but after a rookie season in which he averaged 1.7 blocks per game, that’s never been how he makes stops, relying more on his size than the threat of a swatted shot.

Furthermore, Jordan Clarkson and Tyler Kolek aside, the Knicks’ defense often improves with their bench units, which isn’t common in the league. With Miles McBride and/or Mitchell Robinson out there, the Knicks do more than just hold the opposition to a draw. In the regular season, the Knicks outscored their opponents by 10.6 points per 100 possessions with McBride in the lineup, and by 8.7 points per 100 with Robinson playing.

There’s no doubt that the Knicks have also benefited from some shooting luck in the playoffs. So far, their opponents have hit just 30.5 percent of their shots from three and 73.1 percent of their free throws. Both of those quantities are extremely mean-reverting (defense doesn’t affect an opponent’s three-point success much, or its free-throw percentage hardly at all5). So OK fine, apply the league-wide playoff averages of 34.7 and 76.9 percent shooting instead. That would yield an extra 78 points for the Knicks’ opponents over 14 playoff games, which is meaningful. Even with those revisions, though, the Knicks would still be outscoring their opponent by +13.8 PPG so far in the postseason.

Basically, the math adds up: one big minus (Brunson), one neutral (KAT), and otherwise a bunch of positives sums up to a pretty good defense — if not a little better than that.

It can be a little unorthodox and require high possession-by-possession effort from the wings, however. High floors don’t necessarily imply low ceilings. The Knicks tend to get away with medium effort on both sides of the ball better than most teams, but that doesn’t mean there isn’t a higher gear. It’s easier to find it more often when you don’t run your players ragged with such a heavy minutes load as Thibs did. Hart, Bridges and Anunoby all ranked in the top 5 in minutes per game in 2024-25. That’s a lot to ask of anyone, let alone two-way players who you’re also hoping will be active in transition.

KAT has reshaped the offense

There’s probably been some luck on offense too — the Knicks are shooting an even 40 percent from three in the playoffs, up from 37.3 percent in the regular season. Still, offense and defense aren’t as separable in basketball as is commonly assumed. The Knicks are scoring a spectacular 1.32 points per possession in transition in the playoffs; they actually aren’t getting that many steals, but the ones they do get, largely from their wing defenders, have often been pick-sixes at midcourt that set up automatic points on the other end.

Is anything else different on offense? Well, it’s really just one thing, but it’s major:

They’ve let the KAT out of the bag. (Sorry, I couldn’t resist.) Towns’s assist rate has roughly doubled from the regular season, to the point where he’s a full-on point forward, assisting on shots at nearly the same rate as Brunson. There is a trade-off here: KAT’s usage rate (the percentage of possessions he ends via shots, free throws or turnovers) has fallen, something I wouldn’t necessarily have recommended because he’s a very efficient scorer.

But the critique of the Knicks’ offense one could have raised before is that it was sometimes plodding and predictable under Brunson, even if he’s one of the most effective players in the league late in the shot clock. The Knicks just have a lot more optionality with Brunson and KAT as twin fulcrums in the halfcourt, making better practical use of their spacing. It’s the sort of tweak that elevates the offense from good to great.6

Health matters also, and the Knicks are more or less fully intact. (Though now Robinson has mysteriously injured his pinky finger.) They bulldozed through the Sixers to the point where you almost didn’t notice Anunoby’s absence to close out the series. While the Knicks didn’t have any catastrophic regular-season injuries, Hart missed 16 regular-season games (and played another 13 from the bench), Miles McBride missed half the regular-season schedule, and even Brunson missed 8. While I’m not in love with the deeper part of the Knicks’ bench, it hasn’t been needed as much — many of the minutes have been in garbage time in the various blowouts — and Brown’s “bench mob” mentality gets players who know they’re only going to get a 6-minute stint to go full balls-to-the-wall in those minutes.

The Knicks are more than the sum of their parts

If the Knicks do win their first title since 1973, you might see some recalibration in the conventional wisdom on the much-debated question of “how good must a team’s best player be to win a title?” In April, The Ringer ranked Brunson as the 12th best player in the league. I say that’s too low — I’d take him above Jaylen Brown, for sure, and probably ahead of Donovan Mitchell. But he’s basically a one-way player, and that caps his capacity to compete for, say, the top five. EPM, meanwhile, actually thinks KAT is the Knicks’ best player (!) with Brunson and Anunoby (!!) tied for #2. I’m not sure I totally believe that, either.

On the other hand, I’ve always thought the notion that you must have a top-5 player to win the title is more of a correlation than a causal mechanism. We have a Silver Bulletin tradition of writing about the new NBA champion, and the recurring theme is that the archetypes people have in their heads are too rigid. The Steph Warriors were too young and too small until they weren’t; Nikola Jokic’s defense was supposed to be too much of a liability; the Raptors’ one-year rental of Kawhi Leonard was the sort of thing that had never worked in the past.

Basketball is a team sport, and this is the best team that New York has seen since the 1970s. The front office overpaid for Bridges on the theory that being more than the sum of one’s parts could outweigh mismatched high-end talent of the sort that Cleveland and Philly fans just saw fail. Although the Thibs Knicks were considered overachieving, it’s Brown who has seen the vision through. I’m not gonna lie: Knicks fans are probably going to remember this season fondly even if we’re on the wrong end of a 4-0 sweep. But with a defense that has gone from average to good and an offense that might be legitimately great, the Knicks just might have it in them to find four more wins.

Joseph’s story today is intended as a companion to our PRISM NBA prospect ratings, and PRISM has also been updated with additional data and player commentary since its initial release. Go check out PRISM! -NS

It's easy to forget, given the incredible games we've been gifted this postseason and the carnage that Knicks have caused to the rest of the Eastern Conference, that the most pivotal moment of the NBA season takes place on a random afternoon in May — when deputy commissioner Mark Tatum strides onto a stage with a stack of envelopes, ready to smile while some fanbases realize a year of struggle was for nothing and others celebrate a potential future of perennial contention. Seated in the audience are the draft prospects, the coveted prizes that the front office executives, also in attendance, throw away entire seasons for.

Despite its funkiness, the lottery is perhaps my favorite event of the year. Even beyond my interest in prospect evaluation, it’s one of the few times the NBA honestly focuses on team-building, acknowledging both the decision-making and, yes, the luck required to take a team from sitting in that audience to hosting a playoff game in May. This year’s lottery felt more special to me than any other — I spent an inordinate amount of time on the draft given my work on PRISM, and I felt like I had a vested interest in the order despite my favorite team, the Spurs, not being in the bottom fourteen for the first time since 2019.

The actual TV presentation of the lottery reveal is pretty unnerving. I’m surprised there hasn’t been a Key and Peele sketch on the prospects’ awkwardness as they realize they may have to spend the next five years in Memphis or Utah instead of Miami or Los Angeles. Last year’s lottery reveal had Cooper Flagg making the biggest “this-had-to-be-rigged” face as the Mavericks won the rights to select him #1.

No prospect should feel certain of their destination this year, even if there’s a loose ordering as of right now. My work on PRISM only confirmed this: Cameron Boozer, the projected #3 pick, is ahead in my model by a lot but is unlikely to be selected with the first pick. That makes the 2026 draft unusual — it has so much more top-end talent than a typical class that the team holding the first pick could capitalize and grab additional assets in a trade-down. The lottery results only increased the probability of that happening. The Washington Wizards, who won the lottery, have reportedly been exploring trade options given their comfort level with any of three or four prospects: presumably AJ Dybantsa, Boozer, Darryn Peterson, and Caleb Wilson.

This uncertainty creates diverging scenarios. Depending on your flavor of scouting — I tend to focus more on production, for example — the draft could play out very differently from how I’d run it if I were picking everyone myself.

But that’s what today’s story is intended to highlight. It’s not a “mock draft” or a projection of who teams will pick. It’s who I think they should pick, considering PRISM ratings and team needs.1 Because I’m essentially making myself the GM of all 30 teams, I’ve resisted the temptation to make trades. In real life, I’d trade down if I were the Wizards because I value Boozer more highly than the consensus, but I’m imagining that most of the other 29 Joseph-run teams would also have Boozer #1 on their boards, making deals much harder to find.

The “who they’ll probably take” picks below are based on the consensus mock draft.2 An asterisk by the player’s name indicates the team can’t take their mock-drafted player because they’re already off the board in my draft.

1. 🧙 Washington Wizards

Who they’ll probably take: AJ Dybantsa

Who they should take: Cameron Boozer

The Wizards landed the first overall pick in a somewhat anticlimactic NBA draft lottery, their first win since they nabbed John Wall in 2010. Fun fact: it’s actually Anthony Davis’ THIRD draft lottery win; he was on the Pelicans’ roster in 2019 when they won the Zion Williamson sweepstakes and the Mavericks last season when they picked up Cooper Flagg.

This year’s draft doesn’t have a consensus best prospect the way those ones did, but in the analytics camp, one prospect seems to reign supreme over the others. While there are legitimate gripes with Boozer — he doesn’t project as a rim protector and he gets away with a bunch of travels on his post-ups — it’s hard to ignore this level of dominance from a freshman. He essentially plays as a wingified version of Nikola Jokic, and I expect that to continue as he develops. Yes, I’m aware that’s a lofty comparison, and I’m not saying I expect Boozer to be as good as Joker, but the typecasting of Boozer as a Kevin Love or Al Horford-plus player probably understates his ceiling even if those are reasonable comps.

Boozer will never truly be a poor defender because his strength can help on lighter players trying to drive past him, and because he’ll enter the NBA as the best rebounding wing in the sport. Even if Boozer’s scoring never becomes the main draw of his game, his ancillary skills should be more than enough to make him the best value proposition in the draft. PRISM, I’d note, also focuses on a player’s production over his first seven seasons, when he’s most likely to be on a discounted or otherwise reasonable contract. Sometimes that tilts it a bit toward older players — but Boozer won’t turn 19 until a month after the draft.

The Wizards shouldn’t really be drafting for fit in the first place, but they especially shouldn’t be scared off just because their two best players (Sarr and Davis) are occupying both spots of a double big lineup. Boozer isn’t supposed to be in those spots, projecting more a as an oversized ‘3’, and the Wizards might actually make a playoff push next season if they found a way to incorporate Boozer as a small forward.

Perhaps understanding their optionality, the Wizards have been somewhat secretive thus far on what they plan to do with the pick. They have a budding analytics department and I wouldn’t be surprised if Boozer is on the top of their board and they’re looking for options to trade down, especially since AJ Dybantsa (currently mocked at #1) is coveted by more than a few teams in the lottery this year.

2. 🎷 Utah Jazz

Who they’ll probably take: Darryn Peterson

Who they should take: AJ Dybantsa

It’d be reasonable to assume that the Jazz really want AJ Dybantsa — their owner, Ryan Smith, is a BYU alum, and it’s not everyday that a top draft prospect prefers to live in Utah. In the world where I’m the GM of every team, this works out cleanly — I think Dybantsa is either the second or third best prospect in the class, and the Jazz would look very tantalizing with Dybantsa, Lauri, JJJ, and Walker Kessler in a humongous lineup.

Dybantsa’s combine measurements were super appealing — he measured in at just around 6’9 barefoot, making him a legitimate ‘big wing’, but the concerns about his defense remain. Here’s a disconcerting fact: Reed Sheppard had twice as many blocks as AJ Dybantsa did during their freshman seasons. It’s hard to ignore something like that, even if Dybantsa’s scoring profile is one of the most robust in recent memory. As Jeremias Engelmann wrote, he has some of the signatures of a high-volume, medium-efficiency, middling-defense player who can turn into a quagmire on a bloated contract later. With that said, there are few true can’t-miss prospects on the Flagg tier and PRISM does grade Dybantsa as a “superstar”, a rating that might make him #1 in a weaker draft.

3. 🐻 Memphis Grizzlies

Who they’ll probably take: Cameron Boozer*

Who they should take: Caleb Wilson

Read more

Last updated 6/2/2026

PELE is Silver Bulletin’s rating system for international soccer teams. Each team gets two principal ratings: a PELE rating describing its overall skill level and a Tilt rating indicating its propensity toward attacking or defensive play. Based on these ratings, we can evaluate past match results and forecast future matches. PELE ratings are updated continuously and backdated to 1872 (!).

PELE is also the backbone of our 2026 World Cup forecast. The last section describes the World Cup-specific adjustments we’re making — though “regular PELE” and “World Cup PELE” are about 95 percent the same both in spirit and in practice.

We’re extremely proud of PELE, but it was a lot of work, and it’s not our simplest model. This article describes the system in detail.

Subscribe now

The basics of PELE

PELE stands for Predictive Elo with Lineup Equilibria. We know it’s a little bit nerdy, but this backronym captures most of the essential features of the system:

Predictive means that the goal of PELE is to probabilistically forecast the outcome of future soccer games. These aren’t the FIFA rankings: we’re not interested in which teams are most “deserving” of a particular slot. Rather, we’re looking for factors that have a predictive impact. International football teams play relatively few important games, and some of the most predictive indicators don’t derive from match results alone.

Elo means that PELE shares many properties with an Elo rating system — and indeed, PELE ratings are designed to be comparable to Elo ratings such as the FIFA rankings or the World Football Elo Ratings.1 As with other Elo ratings systems, PELE ratings are updated iteratively at the end of each match, and updates are zero-sum. (If Brazil beats Bolivia 4-1, whatever gain Brazil makes in its PELE rating is offset by a loss of points for Bolivia.) However, PELE deviates from traditional Elo ratings in other important respects, as we’ll describe below.

Lineup means we use player market values and age data from Transfermarkt to help anchor PELE ratings. We look at the market values for the top 23 nationals2 with their respective club teams, with some soft positional constraints. For years since 2005 (when Transfermarkt’s coverage begins3) team ratings are gradually “nudged” toward team aptitude as estimated by these player values. Player ages, weighted by market value, also affect the system — younger teams are expected to improve, while older teams are expected to decline. PELE also uses market values to help calculate whether a team’s strengths are oriented toward offense or defense.

In addition to market value data, we consider each country’s region, GDP, and football legacy. (The regions are not based on FIFA’s six confederations; we’ve developed our own system of 12 overlapping soccer regions, which we believe to be more predictive and more geographically accurate.) But these other factors are less important after the introduction of Transfermarkt data in 2005.

Equilibria serves as a catch-all to capture some other important features of the PELE system. PELE contains many parameters that work together to converge on (we hope) the best possible ratings. But these two mechanisms are particularly important:

• National team results essentially compete with the model’s expectations based on player market values, ages, GDP, region, and team history. The prior gradually pulls each team toward its long-run expectation based on these factors, while the match results push against this if a team consistently outperforms or underperforms PELE’s assumptions.

• PELE calculates two sets of ratings for each team: an Elo-like PELE rating that measures overall squad quality, and a Tilt rating that indicates whether the team tends to be attack- or defense-minded. Tilt ratings are based on (1) whether games involving the team tend to produce more or fewer goals than the model’s expectations, and (2) roster composition. PELE and Tilt ratings can be combined with global scoring trends to derive a score matrix for each game, i.e., the probability that Germany wins exactly 2-1 over Australia, or ties it exactly 0-0, etc. These can be used to estimate win/loss/draw probabilities for any given matchup, or to impute offensive and defensive ratings for each squad.

Scope and coverage of PELE

PELE covers international matches between teams that were both FIFA members at the time. There is an exception for teams that played widely recognized international matches before FIFA was formed in 1904. We think of these pre-FIFA countries as the “Original Ten” (analogous to the NHL’s “Original Six”). They are England, Scotland, Wales, Ireland, Argentina, Uruguay, Austria, Hungary, Belgium and France. Matches between “B-teams” or played under significant roster restrictions (i.e., the Olympics in recent years) are excluded.

Relying on FIFA membership dates essentially outsources decisions about when a national team reached “maturity” to FIFA, an organization we have mixed feelings about. Nonetheless, this provides a semi-objective basis for articulating “official” games among the myriad matches that have occurred over the past 100+ years between 200+ national or sub-national entities. We tested various alternatives to FIFA membership dates, but they only slightly increased the number of PELE-eligible matches while adding more subjectivity. FIFA membership dates are researched precisely.

Nonetheless, we had some decisions to make, particularly regarding which national teams are considered to be continuations of previous teams. Indeed, nearly every political dispute of the past century shows up in some form in historical soccer data. In making these decisions, we tried to be as consistent as possible based on the history, geography and economics of countries being reformulated. FIFA regards West Germany as having inherited pre-WW2 Germany’s football legacy, for example, and considers reunified Germany to be a continuation of West Germany. Our definitions are stricter and treat major changes in national boundaries as discontinuous, such as the split and reunification of Germany, the formation and breakup of the Soviet Union, and the breakup of Yugoslavia.

However, minor changes such as Timor-Leste splitting from Indonesia are tolerated. There are inevitably some judgment calls: we consider the creation and collapse of the United Arab Republic to be a discontinuous event for both Egypt and Syria, for example. In some cases, countries are considered dormant or hibernating and then “reincarnated” if they return to roughly their original boundaries: for example, the Baltic states before and after the formation of the USSR.4 In addition to the 211 current FIFA members, we calculate ratings for 17 nations that were FIFA members at some point but are now essentially defunct, e.g., Czechoslovakia — though defunct teams are excluded from most of the charts we display. Overall, almost 50,000 international matches since 1872 are included for PELE consideration.

PELE Phase 1: “basic” Elo-type ratings

Within our model, PELE is calculated in two phases. The first phase is “simple” and empirically establishes our parameters, such as the regional coefficients or the varying importance of home-field advantage. The second phase introduces mean reversion toward PELE’s expectations based on Transfermarkt player values and other data. This section describes Phase 1.

Some features of PELE are very Elo-like:

• Rating updates are zero-sum. Whenever a team gains ground from match results, this is offset by its opponent losing an equal amount of PELE rating points.

• By definition, the average PELE rating for the 211 active FIFA countries is 1500.

• PELE relies on contemporaneous information. In other words, Hungary’s rating on (for example) April 5th, 2012 is based on information that would have been available as of that date. We don’t go back and recalculate ratings based on post-facto information (i.e. Hungary lost to Norway, but it turns out that Norway was stronger than we assumed at the time).

Any Elo-type system also relies on a number of parameters that govern the overall behavior of the system. To the extent possible, PELE seeks to derive these empirically.5 The next section describes PELE’s approach to some of the most important ones.

Home-field advantage, match importance, and other parameters

We undertake highly detailed calculations for home-field advantage. In general, home-field advantage is very important for international soccer matches, and we think the impact of HFA is underrated by other systems. There are several components to our HFA calculations.

HFA varies over time, and this is derived empirically. In general, HFA increased after WWII, rose until the 1980s/1990s and has been declining since then, perhaps because travel accommodations for visiting teams are improving.

Travel distance impacts our HFA calculations.6 Traveling from Brussels to Amsterdam is less burdensome than flying from South Korea to Brazil. Travel distance is more important for neutral-site games.7

Altitude has a significant impact on teams like Mexico and Bolivia, who typically play their home matches at high altitude. These teams tend to have a significant home-field advantage because they are better acclimated to local conditions. Indeed, the formula is nonlinear. As you may have experienced yourself, engaging in intensive physical activity at 10,000 feet is more than twice as hard as at 5000 feet.

PELE also calculates customized HFA coefficients for each team. More precisely, these measure the spread in team performance, relative to PELE’s expectations, in home versus “nonhome” (neutral + road) games. Like the global HFA rating, each team's custom rating evolves over time. In general, teams in far-flung and war-torn places tend to have larger HFAs, while richer nations in Europe and the Middle East tend to have smaller ones. Bolivia has the largest HFA, mostly because of its altitude. But the customized HFA adjustments are generally pretty conservative; this data is noisy.

Another important consideration is the importance of each match. A friendly will be taken less seriously than a World Cup knockout game. Some previous research, including my own work for ESPN’s Soccer Power Index, suggested that low-impact matches nevertheless provide substantial predictive value. But the actual situation is more subtle. Low-impact matches, such as friendlies, tend to predict performance in future low-impact matches, and most of the dataset consists of these. However, high-impact matches like the World Cup or the Euros tend to better predict performance in future high-impact matches. We considered developing ratings on two parallel tracks (i.e., a friendly match rating versus a “serious” match rating), but it wasn’t quite worth the added complication.8 Instead, we use these importance factors to weight the value assigned to each match. There’s roughly a threefold difference between the most and least important games.

• Friendlies: 0.5-0.7x9 multiplier

• Minor and friendly tournaments: 0.7-0.9x

• Regional tournaments and Olympics10: 0.7x (qualifiers)-1.0x (main tournament)

• Continental tournaments (e.g. the Euros): 1.3x (qualifiers)-1.4x (main tournament)11

• World Cup: 1.5x (qualifiers)-1.6x (main tournament)

In addition, matches that are eligible for extra time (generally, knockout-stage matches) receive an additional multiplier of 1.1. Thus, World Cup knockout-stage matches are treated as having an effective multiplier of 1.76.12

There’s another subtle consideration in PELE: friendly matches tend to slightly collapse the difference in team quality. England will treat a World Cup qualifying match against San Marino with more urgency than a friendly against the same opponent. This is also accounted for by the model.

Finally, PELE ratings include one other feature from Elo’s grounding in chess.13 Most chess ratings systems incorporate a provisional period for new players: their rating updates more rapidly during their initial matches. From an information-theory standpoint, the idea is that the “prior” we have about these players’ skill level is relatively weak, so we ought to be more aggressive about incorporating any data whatsoever from their actual match results. PELE does the same thing. When a new country14 appears in the match file, we weight its initial results more heavily through roughly its 100th international match15, allowing it to converge toward its steady-state rating more quickly.

Differences between PELE and true Elo systems

There are also some important ways in which PELE deviates from traditional Elo ratings.

Some Elo-type systems adhere to a strict principle: winning always helps your rating. In these systems, if England beats San Marino 2-1, England will see some (typically very modest) rating improvement, while San Marino’s rating will decline. The polar opposite is to train the model on the score differential. For instance, if England is expected to defeat San Marino by 5 goals, the 2-1 scoreline will reflect that England underperformed the Elo expectation by 4 goals.

PELE basically strives for a compromise between these. Rating updates are based on what we call harmonic margin16 or “h-margin”. In h-margin, each additional goal has diminishing returns: the second goal in a 2-goal victory counts ½ as much as the first one, the third goal counts ⅓ as much, and so on. This is particularly important in soccer, where the current score substantially affects tactics: if you’re already ahead 3-1, whether to press for another goal or to collapse into a more defensive shell isn’t obvious. Matches won in penalty shootouts — there is some skill in penalties, believe it or not — are regarded by PELE as basically halfway between an outright win and a draw.

Initial ratings and the geography of international football

In Elo-type systems, each team or player usually starts with the same initial rating, typically 1500, before playing any games or matches. However, this can introduce some information loss. To a large extent, which teams are strong at football is predictable. From first principles, for example, you’d expect Argentina to defeat American Samoa. Even if you’d never seen a soccer game, you’d know that Argentina is much larger, has a much longer football legacy, and comes from a region where football plays a much more prominent role in the culture.

Technically speaking, PELE ratings also start out with a blank slate (everyone at 1500). However, we iterate the ratings dozens of times to converge on what we call a “GDP prior” for each country. The GDP prior is based on three factors:

• A country’s purchasing power parity GDP (adjusted for the standard of living) at the time it became eligible for PELE-rated matches. More specifically, we use the natural logarithm of a country’s GDP (there are diminishing returns to economic growth from a soccer standpoint). We use aggregate GDP, not GDP per capita, so both population size and living standards per citizen matter.17 Each country’s GDP is expressed as a fraction of world GDP at the time, so there is no bias introduced from when a country begins playing. Most of the GDP values are taken from the Maddison historical database, which is about 75 percent complete for the countries and years we care about. However, some GDP values are missing in Maddison: for example, Maddison lists the value for United Kingdom GDP rather than for the respective home nations (England, Scotland, etc.) while the home nations compete individually in football. The missing GDP values were filled in using Claude Opus 4.6, with careful oversight from yours truly. For the most part, this process is fairly straightforward. UK GDP can be divided between the home nations based on their relative contribution to the UK’s GDP, for example. Or we can estimate Estonia’s GDP prior to its being taken over by the Soviet Union based on the living standards at the time of comparable countries like Finland.

• A country’s “legacy year”: that is, the first year that it or one of its predecessors became eligible for PELE-rated matches.18 A longer legacy correlates with higher performance even years later, although this is the least important of the three factors discussed here. For the Original Ten, the legacy year is the year of its first widely recognized international match; for all other countries, it’s the year they became a FIFA member. Countries inherit legacy year status from any defunct teams that substantially overlapped with their territory: for example, both Northern Ireland and the Republic of Ireland inherit unified Ireland’s legacy year prior to the Partition of Ireland.

• Finally, we consider a country’s football region. PELE does not use FIFA confederations for anything: these do not correlate all that well with football performance, and can be influenced by political and other arbitrary factors. They can also be blunt instruments for continents as large as Asia. Instead, we crafted our own set of 12 regions, which deliberately contain some overlap. I’m going to be honest, we went through a lot of different versions of these.

Basically, we started out with the six populated continents and then carved out some logical geographic boundaries to create more precision:

• Oceania survives pretty much intact from conventional definitions. Note that Indonesia is considered a transcontinental country, belonging to both Oceania and Asia.19

• The Americas are also self-contained, but the conventional boundary between North and South America doesn’t adequately reflect the cultural distinctions within the region (and how this tends to map to football strength). Instead, we divide the Americas into three regions: 1) North America, defined as the continental territory stretching to the Darien Gap, and 2) the Caribbean, and 3) Latin America. Eligibility for Latin America is defined as any country in the Americas where the predominant language is Spanish or Portuguese. We would have included French also (it’s a Latin language), but there aren’t any FIFA members per se that would qualify on this basis.20 The decision to split off the Caribbean was important. It’s not a terribly important part of the world football-wise. But the idea of the regional groups is that there should be a coherent economic or cultural tie that can inform our priors. And the notion that, say, Haiti’s football performance tells you anything whatsoever about Canada’s or the United States’s rating felt like a big stretch. Even with the 3-way split of the Americas, there is deliberately some overlap: Mexico, for example, is both a North American country and a Latin American country. (We’re sort of implicitly creating a Central America region, in other words, which takes the average of the Latin America and North American values.) Note that Guyana and Suriname are not Latin American countries despite being in South America; instead, they’re considered Caribbean countries.21 Only Canada, the United States and Bermuda22 are purely in North America.23

• Next, we carved out a Middle East region based on its reasonably well-defined geographic boundaries. Most Middle Eastern countries are “taken” from Asia, which badly needs to be subdivided because it consists of so many unlike parts. However, the Middle East also has two FIFA members from Europe (Turkey and Cyprus) and one from Africa (Egypt, which is transcontinental due to the Sinai Peninsula). These countries are treated as hybrids between the Middle East and other regions.

• The other big carveout is Ex-USSR, and we’ll admit that this one is more debatable. The Soviet Union existed for 70 years and affected football culture and development pipelines in ways that are still persistent today. Former Soviet Republics like Moldova are historically weak at soccer as compared to Europe — as is Russia itself, really, given its GDP. But the Central Asian ex-USSR countries are relatively strong as compared with most of Asia. (And Central Asia doesn’t fit neatly into any of our other regions anyway.) The Baltics — Estonia, Latvia, and Lithuania — are considered hybrids between the Ex-USSR and Europe because they were the only former Soviet countries to have football teams prior to the formation of the USSR. We avoid making exceptions based on political developments: Russia is at war with Ukraine, but the war is recent, and Ukraine isn’t (explicitly) a NATO or EU member.

• That leaves Europe as everything in the continent outside the former Soviet Union. We considered some further divisions, e.g., lumping in some nations with ex-USSR into an “Eastern Bloc” region. But these definitions are fuzzy and historically contingent24 and footballing strength has historically been roughly even across Europe, especially once controlling for GDP. Both Western European countries, like England, and Eastern European ones, like Hungary, have strong football traditions.

• Asia and Africa, on the other hand, are large and warrant further division. We already carved out the Middle East and the Asian parts of the ex-USSR region (basically Central Asia) from the rest of Asia. The remaining portion of Asia nevertheless spans a large landmass and more than half the world’s population. The logical division is between South Asia and East Asia. South Asian nations are almost universally underachieving at soccer. East Asia, conversely, contains some of our bigger outliers — notably, South Korea and Japan have become very good at soccer in contrast to the rest of the region. But China is underperforming. Still, priors are priors for a reason: they’re good default assumptions that are sometimes violated. We considered creating an “Asia-Pacific” region that would also include Australia, but this went too far down the road to gerrymandering based on footballing strength. However, Southeast Asian countries are hybrids between South Asia and East Asia in our scheme, so this creates another implicit region along the lines of Central America.

• Finally, Africa is often treated as an undifferentiated mass by Westerners, but nearly all the stronger African sides you’d think of are either in North Africa or West Africa. These countries have different religious and cultural traditions and different relationships with colonial European powers that correlate with football strength. Therefore, Africa is divided into three regions: North Africa, East Africa, and West Africa, with some overlap. North Africa is reasonably well-defined by geographers, so we’re basically splitting the rest of the continent — Sub-Saharan Africa — into two parts. West Africa includes some of the continent’s strongest football countries. East Africa is a slight misnomer: it might be labeled East/Southern Africa if we were even more precise. But teams from that part of the continent don’t tend to reach the same heights.

In case you’re curious, the order of the regional coefficients in terms of how well they predict team quality is:

1. Latin America

2. West Africa

3. Europe

4. North Africa

5. Caribbean

6. East Africa

7. Middle East

8. Ex-USSR

9. North America

10. Oceania

11. East Asia

12. South Asia

This ordering might be surprising, but remember that these ratings control for GDP (and legacy year). Europe might be better than West Africa at football, but its teams have longer histories and it’s much wealthier; the residual coefficient for West Africa is (slightly) higher once you control for that. The Caribbean surprised us as a modestly high-scoring region, but it has many tiny countries that punch above their weight.

PELE Phase 2: advanced ratings with mean reversion and player market values

In traditional Elo-based systems, ratings only change after games/matches are played. Silver Bulletin’s other sports models, like COOPER, already violate this principle because ratings are partly reset or reverted toward priors at the start of each season.25 In international football, however, there’s no “season” per se. So instead, each team’s ratings are very gradually “nudged” toward PELE’s expectations every day based on PELE’s priors. The magnitude of the nudge was calculated by comparing teams’ Phase 1 PELE ratings with their PELE ratings two years later. This allowed us to determine which factors predicted changes in team performance.

However, this is a deliberately slow-moving process: it would take decades for an outlier like Japan to fully converge on its prior. Moreover, teams can push away from this pull toward the prior with consistently good match results. Elo-type systems use a “K-factor” to determine how much the ratings change in response to new results. In Phase 2, our K-factor is set slightly higher than in Phase 1; in other words, match results matter slightly more, giving teams an opportunity to offset the mean-reversion. In practice, the prior is relatively less important for countries that play many international matches, but matters more for teams that compete infrequently in important games.

Prior to 2005, this mean-reversion is just based on the GDP prior: in other words, a country’s GDP, its legacy year and its region. It doesn’t have much effect: regions and legacy years do not change at all, and with some exceptions like China, relative GDPs only change slowly. Thus, prior to 2005, the Phase 2 mean-reversion process barely has a discernible impact on the ratings.

Beginning in 2005, however, an extremely valuable set of data becomes available: player market values and ages as estimated by Transfermarkt. Transfermarkt is exceptionally comprehensive, to the point where we can basically estimate a snapshot of any national team’s market value for any given date since Jan. 1, 2005. These can have a big impact: Norway’s rating is clearly boosted by Erling Haaland’s presence, for instance.

The Transfermarkt data does require some work to process, however. While we could simply add up the aggregate market value for every player of a given nationality, this might introduce coverage bias.26 Furthermore, football matches are 11 players a side, so that France’s 200th-best player is better than Canada’s 200th-best player is irrelevant. Instead, we construct a 23-man roster for each past date. (World Cup rosters traditionally consisted of 23 players, although this has now been expanded to 26.) These rosters are populated in descending order of Transfermarkt values, with a hard constraint on goalkeepers (exactly 3) and some softer constraints on the other positions (a team can’t field a lineup consisting entirely of strikers). The starting 11 gets full credit for its market values, while the reserves (players #12 to #23) receive partial credit based on a sliding scale (the first several reserves get nearly their full market value; the end of the bench doesn’t). If a team doesn’t have enough players listed in Transfermarkt to fill out a 23-man roster, remaining slots are treated as having zero market value.

For the 2026 World Cup specifically, we’ll use actual team rosters rather than our guesstimates once they become available. These will account for injuries and player absences, and players who wind up on different countries than their primary Transfermarkt classification. For “regular” PELE, however, the rosters are constructed algorithmically.

We also compute team ages based on this data, weighting ages based on each player’s market value to their squad. For teams without full rosters, ages are adjusted toward a mean of 26.5 years. Younger teams are usually projected to improve, and older ones to decline.

One big question is whether Transfermarkt data is biased toward certain countries, and particularly toward Europe, since the most valuable club teams in the world are overwhelmingly concentrated there. We investigated this carefully. The original version of PELE made a slightly ad hoc adjustment based on which countries and regions tended to have players who stayed at home even though they would be skilled enough to play for a top-flight European club if they wanted to (for example, this is more common in Latin America than in Africa). But this introduced a somewhat subjective element.

We’ve continued to tinker with the method, but what we’ve settled on this this. In calculating adjustments to Transfarmrket values, we use an average of four regression methods:

• Version #1 applies an empirically-derived penalty to the Transmermarkt valuations associated with players whose club teams are in UEFA. Note that this is based on the club teams of the players and not the team. If ~all of Argentina’s roster plays in Europe, they’ll receive the same penalty that a European club would. On the other hand, some teams like Japan and Mexico will have relatively strong players who opt for domestic leagues; they tend to be helped by this.

• Version #2 instead applies a bonus to players whose club teams are in the Americas or in Asia/Oceania. This is similar to #1 in practice, but note quite identical. The Americas region is the most underrated by Transfermarkt, in terms of predicting future match results, followed by Asia. There is not much competitive club football in Africa but also not much evidence of bias.

• Version #3, however, takes a “kitchen sink” approach with separate regression terms for the Europe, The Americas, Asia/Oceania and Africa. It’s technically the best-performing version of the regression, though it might be slightly overfit.

• On the other hand, Version #4 omits the regional adjustments entirely. It’s basically a hedge or, if you prefer, a version of Bayesian shrinking. We have found some evidence of the Transfermarkt bias declining over time as its coverage becomes even more comprehensive.

Instead, we now just apply a ~30 percent discount to player values as listed on club teams in UEFA nations. This corrects for Transfermarkt’s modest pro-Euro bias and is more accurate empirically.

As of May 28, we also added another adjustment to the Transfermarkt ratings that accounts for what share of a team’s players participate in domestic club leagues. There is a small amount of signal here — having your players stay at home can be a sign of a solid pipeline and might create more chemistry for the national team — but it’s quite minor.

There’s one other subtle factor in calculating our historic ratings. Suddenly, a whole bunch of new data becomes available on 1/1/2005. Rather than easing into the new regime gradually, we found PELE performed considerably better in the early Transfermarkt era (~2005-2010) if we made a one-time step-function adjustment to team ratings on 1/1/2005 to account for the new data. Essentially, this step-function banks in 10 years of reversion toward the new, more informed prior. If you look very carefully at PELE’s historic ratings, you may see bigger changes in 2005 than in other years.

However, as valuable as the Transfermarkt data is, the “Phase 1” PELE ratings are already pretty smart based on match results and the GDP prior.27 The player data is certainly worth worrying about and aligns PELE with betting odds more precisely, but it’s an important factor rather than a dominant mechanism in the system.28

Tilt ratings and expected goals

So far, I’ve barely even described one of the most important features of our system: Tilt ratings.

When I created SPI for ESPN for the 2010 World Cup, our system had separate offensive and defensive ratings for each team, characterized as their projected number of goals scored and allowed per match. The offensive and defensive ratings could then be combined to project scores and calculate overall quality ratings.

While I think SPI was a smart system, I actually think this technique wasn’t well-suited to soccer. Unlike a sport like American football, offense and defense are fluid in the sport. There isn’t a distinct platoon on either side. And soccer is deeply tactical: a team might play completely differently with a 1-goal lead than a 1-goal deficit. A game between two quality opponents could easily turn into a tight match or a shootout.

However, we can evaluate which teams tend to be involved in high-scoring games. This is basically what Tilt does: it measures whether matches featuring the team tend to involve more combined goals for both sides (positive tilt) or fewer (negative tilt). So while PELE is our measure of overall team quality, Tilt is more a measure of mindset. In sporting gambling terms, you’d use PELE to set the point spread or the odds of a team winning, and Tilt to project the over-under.

However, having a positive or negative tilt rating isn’t inherently good or bad. Teams can succeed — or fail — with more attacking styles or more defensive ones. In fact, PELE and Tilt ratings are, by design, mostly uncorrelated (the cool kids would say they’re orthogonal):

We can, however, combine PELE and Tilt to project goals scored and allowed in each match. We do this in the round-robin table of the PELE landing page, for example, which simulates a round of matches on a neutral field between all of the 211 FIFA teams and all 210 of their opponents.29 A team’s projected number of goals scored and allowed against the round-robin are basically equivalent to SPI offense and defensive ratings, just with the process reversed:

• SPI: Offensive rating + Defensive rating → Overall quality rating

• PELE: Overall quality rating (PELE) + Tilt → implicit offensive and defensive ratings

This process can also be used to project the odds of a win, loss or draw between any two teams under any circumstance (home, road, neutral, etc.). Indeed, this is the process we’ll use for our probabilistic World Cup projections. But it is among the more complicated aspects of the system.

Our starting point is to create a projection of the number of goals in the match, derived from our database of nearly 50,000 historical results. The ingredients in this projection are as follows:

• By far the most important factor: our rolling leaguewide baseline of overall goal-scoring in typical matches. Soccer has gone from featuring 4-5 goals per game at its inception to just 2-3 goals (combined between both teams) per match now. Our model calculates this baseline by averaging the past 5 years of data from all international matches, with a trendline term (is overall goal-scoring rising or falling globally?) and a correction that downweights outlier matches (Australia 31–0 American Samoa).

• We also account for the difference in team quality, as measured by PELE.30 Matches between teams with large rating gaps tend to yield much higher scores.31

• The importance of the match: more important matches tend to play tighter and feature fewer goals. In general in soccer, higher quality of play is associated with fewer goals: it’s completely different from something like the NBA in this regard.

• And whether the game was played at a neutral site; neutral-site matches tend to play a little more wide-open.32

Next, we looked at whether there were systematic differences in scoring propensity, i.e. whether some teams tend to produce higher-scoring games beyond what their PELE rating might imply. Our Tilt ratings are the solution to this: Tilt is basically the difference between actual goals and expected goals based on our formula (regressed strongly toward a mean of zero because goals are rare in soccer and the raw signal is noisy). If matches involving Germany (canonically attack-minded) tend to produce higher scores than others, Germany will get a positive tilt rating, while a negative rating indicates a team whose matches tend to produce lower-than-expected scores like Senegal.

Tilt ratings have two subcomponents. Since Transfermarkt player data begins in 2005, we can sum up all the player values and allocate them to either offense or defense based on their positional assignments. As of mid-2026, for example, most of Norway’s value is concentrated on offense (Haaland) while most of Nigeria’s is on defense. The overall split is designed to be 50/50 based on these positional allocations:

The resulting effect on goal-scoring is what you might predict: teams whose strengths are concentrated on offense tend to produce higher scoring games.

The second component is tactical tilt. Figuratively, it’s whether a team prefers to play a more open/attacking style or a tighter, more defensive one, trends that can be persistent over decades based on the coaching regime and the soccer tradition in each country. More literally, it’s the residual number of goals scored relative to PELE’s baseline expectations.33 Tilt ratings are strongly hedged toward a mean of zero since measuring this attack/defense tendency is noisier than measuring differences in team quality.

The score matrix and future match predictions

From PELE and Tilt, we can project the number of goals for each team in any given match. Technically, the way the model does this is by first projecting overall goals based on leaguewide trends and each team’s Tilt rating, and then dividing the goals between the projected winner and loser.

But a projection like Spain 2.7-Finland 0.8 only tells you so much. Soccer is a low scoring game with a lot of draws, so the precise number of goals matters. Spain can score exactly 2 goals or exactly 3 goals or zero goals or some other integer, but they can never finish the game with 2.7 goals.

The Poisson distribution is designed to handle this sort of situation34 and is the traditional choice in soccer models. But it tends to have some problems. Particularly, Poisson understates variation too much, tending to underestimate both the number of draws (especially 0-0 draws) and the number of blowouts (Germany 7-1 Brazil). PELE’s solution35 is a negative binomial distribution with a correlation term. It’s not so important you know precisely what this means but that you capture this basic intuition: one team’s score affects the other team’s tactics. A 0-0 match tends to play tighter 75 minutes in than 3-2: it’s already been a low-scoring game but it will often play tighter still from that point forward. Our method does a good job of matching empirical goal-scoring distributions: e.g., there are about the right number of draws and blowouts in PELE.

PELE calculates a precise score matrix for each game (the chance Team A beats Team B by a score of exactly X-Y). It then sums up the cells in this matrix to estimate probabilities that the game ends in a win, loss or draw. For instance, here is the matrix for the United States’ first World Cup match against Paraguay on June 12, projected to be a low-scoring affair:

There are some complications introduced by extra time and penalty kicks, such as will be used in the World Cup knockout stage. Games eligible for extra time potentially increase the amount of gameplay by roughly 33 percent and empirically convey a slightly heavier advantage to favorites. PELE accounts for all of this. Historically, about 40 percent of draws after regulation are resolved in extra time before penalties, so PELE takes 40 percent of games projected to be draws in regulation and assigns a winner.36 The other 60 percent of extra-time games go to penalties.

Based on our analysis of several hundred past penalty shootouts, there actually is some skill in them; better teams and home teams tend to win shootouts more often, though edges in most situations are rarely more than about 60/40.

World Cup adjustments

PELE’s projections for the World Cup are based on a highly similar process to the one already described here. PELE already “forecasts” past games to help calibrate its ratings, and makes projections of future matches as described above.

So most of the work is just in applying the PELE process to the structure of the World Cup’s group and knockout stages. We very precisely simulate the group stage and knockout stage, typically 100,000 times per model run, including calculating each official tiebreaker and FIFA’s complicated process for assigning third-place teams to knockout-stage slots. Actual game results will gradually replace simulated ones as real World Cup matches are played.

There are, however, a few small tweaks we’ve made that only apply to our World Cup projections and aren’t used in regular PELE:

• Instead of constructing a roster algorithmically, we select the best 23 players from among a team’s official 26-man World Cup roster. Mainly, this accounts for injuries, and players who are eligible to participate for multiple countries and where the default Transfermarkt nationality puts the player on the “wrong” squad. We examine the difference in Transfermarkt value between the algorithmic roster and the World Cup roster and apply a bonus or penalty to each team accordingly, based on our analysis of past World Cup rosters since 2006. In general, the differences are not all that large. In most cases, teams lose players that the algorithm would prefer rather than gain players who add more value. However the adjustments are fixed to be zero-sum across the 48 World Cup entrants so roughly as many teams gain PELE rating points from this process as lose from it.

• We also account for injuries and disciplinary suspensions on a match-by-match basis, using publicly reported injury information. There is inherently some guesswork here in translating ambiguous media reports into predictions about who will and won’t be available for a given match. However, often these injuries are less impactful than you might assume because World Cup rosters are typically very deep. Teams also have some wiggle room since only the best 23 players out of the 26-man roster are used in calculating roster value.

• For the third group-stage match day only, our World Cup simulation considers team incentives37:

  • If both teams would advance with a draw (but have not otherwise clinched advancement), they have strong incentive to essentially collude to keep the score low. While there are not many matches like this in the historical data, they’re quite low-scoring. The model reduces expected scoring by 1 combined goal in these games.

  • Conversely, if both teams need a win to have a chance to advance (but have not already been mathematically eliminated), teams have incentive to go for broke, and expected goal scoring is 1 goal higher than in the default PELE formula.

• PELE simulates the expected number of bookings (yellow cards and red cards) for World Cup matches because this is used as part of FIFA’s Fair Play Points tiebreaker. (Fewer bookings is better.) These are not random: better teams tend to receive fewer bookings, but teams receive more bookings when against tougher opponents. Home teams receive fewer bookings, an advantage for the U.S., Canada and Mexico.

• As in regular PELE, team ratings, including Tilt ratings, are updated at the end of each game based on a comparison of actual and expected h-margin. In other words, we run the simulations “hot”. Teams have an opportunity to overperform or underperform their pre-World Cup rating based on the results of both actual and simulated matchups, and this carries forward to the rest of the tournament. For instance, if Belgium wins 5-0 in its first group-stage match in simulation #12616, its rating will be revised upward for the rest of the tournament within that particular sim — or if Belgium wins 5-0 in its actual first match, its rating will improve for the rest of the tournament in all simulations. We use a slightly higher k-factor for our World Cup simulations than in regular PELE to account for short-term fluctuations in “form”. There is some evidence in the historical data that teams are slightly streaky over short periods.

• There is an empirical tendency for World Cup group-stage matches to be more upset-prone than the default PELE formula expects, but knockout-stage games to be more “chalky” with favorites winning more often. We aren’t quite sure about the mechanism for this: it may be teams feeling “nerves” early, for instance, or group-stage incentives (often a draw is a good result) compressing team quality to some extent. Thus, our World Cup simulations account for this, applying a 0.9x multiplier on the difference in PELE ratings to group-stage matchups but a 1.1x multiplier to knockout-stage games. This makes a relatively small difference, but it means that the forecasted match projections listed on our World Cup page will not exactly match the ones listed on the PELE page.

• PELE uses some very slow-moving league calibration factors, such as the total number of goals scored per game in competitive matches across global football and the overall default magnitude for home-field advantage. These generally take years to change, and so we just freeze them in place for the roughly month-long World Cup.

• PELE adjusts for altitude as part of its home-field ratings: playing at altitude typically offers a significantly larger advantage to the home team. In regular PELE, we use a default altitude for each country. For the World Cup, we use the altitude of the specific venue for the match. Thus, Mexico has a larger HFA in Mexico City than in Monterrey, for example. Note that, despite teams often arriving at the World Cup venue weeks in advance to acclimate, we haven’t found any evidence that home-field or travel factors are any less important in the World Cup than in other tournaments.38

All other factors are the same. Knockout-stage matchups include simulations of extra time and (if necessary) penalties, but this is also true for regular PELE. We thought about building in an adjustment for rest days, but FIFA actually does a pretty good job of building in adequate rest between matches for each side.

Subscribe now

If you’re an avid listener of the Bill Simmons podcast1, you might have heard a recent bit he did with Zach Lowe, where they pondered whether this was the NBA’s weirdest season ever. For what it’s worth, excluding global pandemics, that claim has a lot of merit. The Clippers started the season near the bottom of the league amidst allegations they paid Kawhi Leonard 28 million dollars under the table; a gambling scandal led to the arrest of a coach and a player on the third day of the season; Bam Adebayo scored 83 points in a game; and fully a third of the league was literally trying to lose games.

It’s hard to say whether the weirdness has undermined the NBA’s effort to rebrand the league in a forward direction. With the league moving most of its games to NBC and Amazon Prime, the presentation has focused less on its legacy stars — Steph, Durant, LeBron — and more on its insurgent young core. The guy everyone wants to watch in the playoffs is Victor Wembanyama, not LeBron. But the shift has been backed by an influx of talent. I spent a lot of time analyzing this while building out our NBA draft model, PRISM, and it was hard to miss just how much talent has entered the league over the last few years.

This year’s Rookie of the Year race has been one of the best arguments for that explosion of talent — and, fittingly, one of the weirder ones we’ve seen in a while. Kon Knueppel and Cooper Flagg, former Duke teammates and roommates, traded the #1 spot on NBA.com’s Kia Rookie Ladder all season long. Flagg, the No. 1 overall pick (and the namesake for our college basketball model) has the highlights — 51 points against Orlando, a 42-point game in Utah before he turned 19, the kind of explosive, load-bearing performances that typically lock up the award. Knueppel, taken three picks later, has been the quieter storm: he leads the league in three-pointers made while shooting 43 percent from deep, and played a leading role in Charlotte’s leap from 19 to 43 wins. Flagg is ahead in counting stats per game across the board, but Knueppel has the efficiency and wins.

Subscribe now

Is the stats case even close?

There have been attempts to flatten Knueppel's case into some version of "he shoots threes really well," which ignores the off-ball gravity, passing, defensive IQ, and the way he's made Charlotte's entire offense function differently. Flagg’s case is also bigger than “he’s dominating the box score on a bad team” and people should be accounting for poorer context — the chaotic situation in Dallas — which makes it harder for him to produce at the same efficiency as Knueppel.

This framing is not so uncommon in awards discussions — the player with higher counting stats against the player with higher efficiency — and this type of discourse tends to get fans riled up, which is probably why mainstream outlets love to play into it. Analytical types tend to frame basketball around impact, which cuts straight through the noise of trying to calibrate all of these stats. While I don’t think ranking players is as cut-and-dried as just taking some aggregation of their impact metrics, there isn’t really an argument for Flagg from a pure stats point of view. On a per-possession level, he lagged far behind Knueppel.

Voters aren’t answering the same question

Of course, there is an argument against advanced metrics here that isn’t just luddite pessimism — stats like EPM, part of a family of RAPM-type impact stats, don’t always do the best job of distinguishing between a player’s impact and the difficulty of their role. In fact, just the opposite: RAPM and its descendants measure how team-level point differential shifts when a player is on vs. off the court and implicitly tend to reward being on good teams. That’s genuinely useful, but it tends to measure how good a player is at the thing they’re currently being asked to do — as opposed to either how rare the thing they’re being asked to do is, or what they might be able to do in another context.

Knueppel’s role in Charlotte — moving without the ball, spacing the floor, catching and shooting threes within a functional offensive system — is considered an easier adjustment in the NBA. He averages under two dribbles per touch — whereas Flagg’s role in Dallas, which involves a lot of primary creation, high-usage shot generation, and running the offense as an 18-year-old on a team with no other reliable engine, structurally suppresses the efficiency numbers these models care about.

For what it’s worth, Rookie of the Year voters traditionally do care about how much of an advantage creator you are. Knueppel would have the lowest on-ball percentage of any Rookie of the Year since Karl-Anthony Towns won the award in 2016.

So, yeah, EPM can tell you that Knueppel's minutes are technically more productive than Flagg's, but that doesn't settle the vote the way some analytics people want it to. The strongest version of the Flagg case isn't really about counting stats or highlights. Rather, it's a reframing of what the award is measuring — that Rookie of the Year should go to the player who has adjusted to the NBA best, and under that lens, Flagg's case sharpens: he's 19 years old, carrying a bad roster as its primary creator, producing against elite defenders scheming against him every night, and doing it at a level only LeBron, Luka, and a handful of other all-time teenagers have matched.

Still, that framing is building in some implicit credit for potential — Knueppel would be a very valuable player with or without rookie status, while we can’t definitively say that about Flagg this year. And even if we adjust for playtype difficulty, Kon is still ahead of Flagg in efficiency.2

But some Rookie of the Year voters aren’t necessarily even answering the question of which rookie was the best player so much as taking their mandate to mean “who will be the best player?” — and under that criterion, Flagg takes the cake.

Advanced stats like eWINS are admittedly not for everyone. They expose some inconvenient truths: rookies usually stink on defense, and their shiny, high-volume box score production is often paired with middling (at best) efficiency. The arrangement is a win-win — the player gets seasoning and sneaker deals, while their teams remain comfortably within the lottery — so long as you don’t ask too many questions of the data. And most voters don’t. Jae’Sean Tate, for example, had the highest eWINS in 2020 and got zero first-place ROTY votes — most of his value came on the end of the floor voters aren’t paying much attention to.

This year’s race only halfway matches that template. Kon and Cooper have been roughly similar defensively by most advanced metrics — and both are rated as about league average, which is unusual and promising for a rookie. Yet Flagg carries the reputation of an elite defender because without enough signal in the box score, the media is defaulting to their priors about his long-term defensive projection.

There’s also a more subtle issue: when voters treat ROTY as a forward-looking judgment rather than a full-season retrospective, they naturally time-decay the regular season. Victor Wembanyama was clearly better than Chet Holmgren by the end of 2022-23, and voters treated that midseason leap as license to wipe away his early-season struggles.

All of this would be forgivable — maybe even commendable — if voters were actually good at picking the best future player. The problem is they aren’t: take a look at that table above. Tyreke Evans over Steph Curry? Or even Kyrie over Kawhi? ROTY voters have often systematically underrated the lower-drafted guy.

The prediction markets won't sit still

But what makes the race genuinely weird is a string of late reversals in the conventional wisdom — and an unexpected twist involving former ROTY Luka Doncic. When I first planned on writing about Knueppel, I was confident he’d more or less locked the award up, and as of a few weeks ago, the prediction markets agreed, giving him a 94 percent chance at the trophy. Then Flagg put together back-to-back statement games, and Knueppel’s odds cratered to as low as 21 percent.

Silver Bulletin has covered prediction markets extensively in the past.3 They’re often very smart, and they’re certainly not easy to beat. But as Nate wrote about the election of the new pope last year — Leo XIV had been trading at only 1 percent odds — they aren’t necessarily at their best when trying to anticipate the behavior of a cloistered group of insiders. Polymarket’s ROTY contracts have millions of dollars in volume, which don’t exactly approach something like its presidential election markets, but are also not the kind of market you’d expect to flip 50 points in a day.

The first thing worth flagging is that we actually do have information about how voters feel. Only a few days before Flagg’s big weekend earlier this month, ESPN conducted a straw poll of 100 media members, some of whom are in the pool that actually votes on the award. Knueppel received 80 first-place votes to Flagg’s 20. So we have a prior suggesting as of just two weeks ago, Knueppel had a decent lead over Flagg — a lead hard to eclipse even with a full week of regular-season dominance.

Tim Bontemps, the reporter who conducts the straw poll, said something interesting on the Hoop Collective podcast the other day — that an 80-20 split in a room of 100 voters doesn’t mean each individual voter is 80 percent sure. It’s more like a collection of 60-40 decisions that mostly broke the same way. That’s a good point, but let’s analyze it. The lower bound on what 80-20 means is that every single Knueppel voter was close to 51-49 and just barely tipped his way. The upper bound is that all 80 were completely certain. The vote count alone can’t distinguish between those worlds, and Bontemps is right that we shouldn’t confuse margin of victory with depth of conviction.

But we can actually test what happens between the straw poll and the real thing, because Bontemps has been running the MVP version of this poll since 2017. In seven seasons of MVP straw polls, the leader has won the actual award every single time. In 2021-22, Nikola Jokic’s straw poll lead of 62-29 widened slightly to 65-26 in the real vote. In 2020-21, his 89 percent straw poll share held almost exactly at 90 percent. The two most recent seasons did narrow slightly — Jokic went from 85 percent to 80 percent in 2023-24, and SGA went from 77 percent to 71 percent in 2024-25 — but neither came close to flipping.

The bull case for Flagg probably comes from 2022-23. That year, Joel Embiid led the straw poll, but it was essentially a tie: he actually had two fewer first-place votes than Jokic. When the actual ballots came in, Embiid won 73-15 in first-place votes. Despite the close straw poll, voters didn’t scatter — rather, they broke hard in the same direction.

Steve Aschburner, the NBA.com writer who runs the Kia Rookie Ladder and is himself a voter, published his final ballot on April 8th with Knueppel at No. 1, and directly addressed the weekend performances that swung the odds: the 96-point two-game stretch swayed oddsmakers, he wrote, but didn’t eclipse Knueppel’s body of work. Historically Aschburner’s Rookie of the Year ladder has correlated strongly with the actual vote share — his one recent miss being Evan Mobley over Scottie Barnes in 2022, which was decided by 5 first-place votes, the narrowest under the current format. Bill Simmons has also rallied around Knueppel. Two voters, neither of whom are necessarily advanced-stats truthers, have come out with their vote, and they’re sticking with Kon.

That raises another issue: Could there be some form of insider trading? It’s obviously not an out-of-bounds question. Still, the scope for it might be limited on this market — no single voter’s private knowledge of their own ballot moves the needle much in a 100-person electorate, and any conspiracy between media members to collude on voting would be difficult, especially before the ballots come out. Building sharp models in awards markets is also difficult — these are one-off events with small voting bodies, so there’s no clean way to build a model for an event where the criteria can shift with narratives and storylines.

The market has corrected multiple times since the initial flip, with Knueppel briefly reclaiming favorite status before Flagg snatched it back over the last few days. Kon didn’t help his case in the Hornets’ dramatic play-in win on Tuesday — he had one of the worst games of his career, capped by a late-game benching. Despite being the NBA’s leader in threes as a rookie, he was absent from the floor over the last seven minutes, even when the Hornets desperately needed a three to send them into overtime.

And yes, even though the play-in game isn’t technically part of the regular season — statistically, it exists in purgatory, since it isn’t part of the postseason either — it’s going to bleed into voter perception. Because of Luka Doncic’s minimum games appeal, ballots were delayed, and Knueppel’s performance on Tuesday might empower some of those wishy-washy 51-49 Kon voters to go with Flagg, even if that’s technically not within the criteria of the award. The market certainly believes that voters will be swayed — following Kon’s game, Flagg is now inching toward being almost a 3:1 favorite.

The prediction market feedback loop

It would be easier to map out this race if close Rookie of the Year races happened more often. Over the last twenty years there have only been two nail-biters — Scottie Barnes over Evan Mobley in 2022, which came down to a 15-point margin and is still the closest vote since the current format started, and Evans over Curry in 2010.

Everything else has been a blowout, or close enough to one that the discourse around it never really mattered. Wemby was unanimous, KAT was unanimous, Lillard was unanimous, and even the races that felt close at the time, like Ben Simmons and Donovan Mitchell in 2018, turned out not to be. Perhaps the closest precedent is 2007-08, when Kevin Durant took home the trophy despite significantly worse advanced stats than Al Horford, who helped lead the Hawks to the playoffs as an 8th seed. That would seem like bad news for Knueppel, but the voting body is different now, in that they’re more receptive to the use of advanced stats than in 2008.

So, historical analogue or not, when people look at the current market and say it's behaving weirdly, part of what they're picking up on is that we don't really have a reference class for what a close ROTY race is supposed to look like, especially between two players with such different profiles. Some of what looks like noise is just what happens when a binary outcome gets close enough that small inputs start to matter: every new game has more leverage in a way that doesn’t seem rational on the surface.

Of course, that’s really just another way of saying buzz is dominating a race that, on the merits, might not be that close. And maybe this is where the whole thing gets a little recursive. The “buzz” isn’t just a reflection of the race, it’s also an input into it. On one side you have a player who’s had a great rookie year, and on the other, a guy who looks like a future superstar, even if he was an outright negative earlier in the season. Those are different kinds of cases, and voters trying to be rigorous about their ballot are still going to get nudged by which story feels more compelling in a given moment. A 30-point Flagg game in March probably carries different weight than a 30-point Kon game because of what it implies about the next 5, 10 or 20 years.

Prediction markets potentially feed into the same loop. They’re aggregating information, but in a race like this, most of what they’re aggregating is the buzz itself — which podcasts said what, which highlights went viral. You know, the vibes. It’s not that prediction markets are broken per se, it’s that there’s just not much independent information for them to work with — so they end up measuring the zeitgeist and handing it back with the authority of a price.

Now, do I think Kon is still a good bet at 28 percent? I think so, because the vote trackers show a fairly close race, and the priors for him are pretty strong. Still, the ballot delay gives recency bias more room to breathe than it usually gets, and if Kon has another rough night tonight it’s going to get harder to shrug off his doubters. Two weeks ago I would’ve told you this race was over. While the underlying evidence from an 82-game season hasn’t changed very much, if Flagg is holding the trophy in a few weeks, I will no longer be surprised.

Don’t get me wrong: I’m excited about the NBA playoffs. But my interest in the NBA regular season has been flagging in a way that it hasn’t in a long time. There’s a simple reason: it’s the tanking. A full third of the league — five teams in each conference — basically gave up on the season at some point between October and February. The identities of the 10 playoff and play-in teams in each conference were practically locked in a few weeks ago — and they’re literally locked in now.

The NBA is acutely aware of the issue, though it’s unclear whether the league considers it a real problem or just a PR issue. (Here’s why it’s an actual problem: about half the games on any given night “feature” a tanking team. I’m not about to pay $200 a ticket to see a team that isn’t even trying to win play the Knicks at MSG.) And the solutions it has proposed are mostly tinkering around the edges with the current rules, full of the same kinks and quirks that will be exploited by future Sam Prestis and Daryl Moreys.

Maybe that’s because it’s not quite clear what the league wants. Losing isn’t the same thing as tanking, and perhaps it’s fine to reward the former but not the latter. Up to a point, you might want to help teams for sustained losing as opposed to a one-year fluke like this year’s Indiana Pacers, but maybe not if they’re perpetual basement-dwellers. Other things equal, you probably don’t want the same team to get a #1 or #2 pick several times in a short period or for a potential contender to luck into a top four pick. Meanwhile, the myriad of pick swaps and trade conditions creates weird cliffs in the system and perverse incentives. And while there’s a lot of focus on losing teams, it’s also not clear you want winning teams to be able to trade draft capital seven years out when ownership and management may turn over anyway once it’s time to pay the piper.

So what if I told you there’s an alternative that could accomplish all of the following?

• Introduce explicit penalties for tanking or for repeatedly failing to reach even the play-in game.

• Substantially simplify trade rules.

• Flatten the lottery odds to the extent desired by the league, especially toward the top.

• Protect teams against themselves by limiting the amount of future capital they can trade.

• Eliminate cliffs in the system, such as when a team would have to give away its pick if it climbs above a given position in the standings.

• Reduce the element of luck: bad teams would still be rewarded with better capital, but they wouldn’t be quite as subject to the literal bounces of the lottery balls.

• Reduce the chance of already good teams landing top picks.

• Put some curbs on the same team repeatedly winning top picks.

• Provide teams with more flexibility to suit team needs, such as carrying over capital from season to season or acquiring depth rather than a single highly touted prospect who isn’t a good fit.

• Usually guarantee a team the opportunity to call its shots if there’s a player it really likes in the rough vicinity of its board.

• And make “draft night” — now auction night — even more strategic and fun.

Sounds pretty good, right? But it does require one leap of faith. As I’ve been teasing at, we’ll be getting rid of the “draft” per se. No, I’m not advocating for all players to simply become free agents, although there are worse ideas. I don’t think it’s inherently a bad idea to “redistribute the wealth”. But every fantasy football nerd’s favorite solution, an auction rather than a draft, provides for a lot more flexibility to tweak the knobs to the league’s desired levels.1

Subscribe now

True, you could accomplish some of these things within a draft format. But once I started down this path, the auction route proved far more flexible. For instance, you can penalize teams in a proportionate way for tanking by deducting a little bit of capital instead of either giving them a slap on the wrist or employing a “nuclear option” like taking away draft picks entirely.

The basics of ARC

The gist of the system is this: the draft would be replaced with an auction-type system, where teams bid on eligible players each June with Acquisition Rights Capital or ARC. I thought about calling this something cheeky like “DraftBucks”, but a) there’s not really a draft any longer, and b) I wanted to go with the sort of technical term that could plausibly appear in the CBA. ARC, of course, can be traded, but the mechanisms for doing this are simpler than under the current draft format. A unit of ARC can alternatively be designated with an arc symbol ⏜.

Needless to say, the devil is in the details for any proposal like this. There are 21 rules governing the use of ARC. You could tinker with most of them without really messing with the overall spirit of the system, however.

Let’s start with the two most important rules:

1. At the end of each season, teams receive between 25 and 100 ARC based on their order of finish. As under the current system, playoff teams are slotted 1-16, (though the two NBA finalists would be ranked first and second2) followed by non-playoff “lottery” teams.

2. There are no more swaps or conditional picks. Instead, all trades involving future asset rights are made with ARC. All trades or player acquisitions involving ARC must be made in whole units; ARC cannot be fractionalized.

Rule #2 is, I hope, relatively self-explanatory. ARC is fungible, so there’s no benefit in trading, say, Miami Heat ARC for Minnesota Timberwolves ARC. So, instead of “the Rockets acquired Giannis Antetokounmpo for Reed Sheppard, Jabari Smith Jr. and [insert long and convoluted description of picks and swaps]” the transaction wire would read as “the Rockets acquired Giannis Antetokounmpo for Reed Sheppard, Jabari Smith Jr. and 50 ARC”. There’s actually a lot less to keep track of.

As for the allocation, here’s what I had in mind:

There are some subtleties here. Under the current system, the steepest part of the curve, in terms of the expected value gain from dropping an additional position in the standings, is roughly between the 20th and 23rd positions. For instance, the 20th-ranked team has just a 9.4 percent chance of lucking into a top-four pick; that nearly triples to 26.3 percent for the 23rd-ranked team. Teams in this region can usually play a little bit and have a plausible chance of at least making the play-in round. But the NBA’s current setup strongly deters them from competing; our ARC system would reduce the marginal gain from losing in this region of the curve by roughly half. Instead, under ARC, the steepest part of the curve is for teams that have made playoffs anyway; it’s flatter toward the top and the bottom. And the ARC allocation for the three worst teams would be completely flat, similar to how the league awards the same 14 percent chance of receiving the #1 overall pick to the bottom three teams now.3

Here’s how I estimate what each draft slot is worth. The metric we use to evaluate player outcomes in our new NBA draft model, PRISM, is based on Estimated Plus-Minus (EPM). We tallied the number of wins a player produced above replacement level in his first seven NBA seasons (when he’s most likely to be on some sort of cost-controlled contract with his original team4), rounding below-replacement-level seasons up to zero. I then drew a smooth logarithmic curve over this data based on a mix of the average WAR and the median WAR historically associated with each draft slot. Overall, we calculate that ~55 players in each draft class5 produce enough expected wins to be worth a guaranteed contract, which is a close empirical match for how NBA teams behave in reality.6

Under our system, there are 1860 ARC distributed to teams in any given season. (As we’ll get to later, teams can have their ARC allocation penalized under some circumstances, but any “taxes” the league collects are redistributed into the system.) If we divide ⏜1860 by the expected value above replacement level associated with each draft slot, we come up with the following:

As you can see, draft value becomes nonlinear toward the top: the #1 overall pick is worth about twice as much as the #7 pick. Having looked at a lot of other “draft charts”, I’m comfortable that’s somewhere in the right ballpark.7 But value tends to flatten out once you get out of the lottery. The first half of the second round can produce tangible value — sometimes even a lot of value (Nikola Jokic was drafted at #41). But the back half of the second round literally zeroes out; teams are usually indifferent toward keeping the players they select at all.

The chart does hint at one important feature of the system: there’s a ⏜100 cap on how much a team can bid on any given player. If several teams make the same bid, then the league goes back to the ping-pong balls to determine who gets him. So there is still some element of luck: typically, in any given draft, the top three picks will be worth more than the max 100 ARC allocation. You can’t just stockpile ARC and guarantee yourself a Wemby or a Cooper Flagg, in other words.

We’ll return to some of these details later, but first let me run one more comparison between ARC and the existing rules. This chart shows the expected value of each draft slot under the current system based on the lottery odds for first round picks and compares it to each slot’s ARC allocation under our new model. Importantly, the expected value estimates also include second-round picks, which provide a sneaky benefit to the worst teams since the first half of the second round is worth a lot more than the back end. (Under ARC, there’s no more second round because there’s no more draft; teams just keep bidding on players until certain criteria are exhausted.)

So yes, we’re taking a little bit from the poor and giving it to the rest of the league. Finishing in one of the bottom eight positions in the standings is now worth slightly less, especially in the bottom three. But it isn’t that big a shift, and the largest benefits convey to the league’s middle class.

Plus, even if we’re giving them less capital, we’re probably reducing variance for the worst teams. A typical draft features three players who are worth a ⏜100 max bid, and three teams each season who are guaranteed to receive 100 ARC. Surely, some other teams will acquire the capital for a ⏜100 bid via trade (or because they have some ARC left over from previous seasons; there’s a mechanism for that). But generally, teams will have more control over their own destiny. If there’s a player they really like — say, Steph Curry in 2009 — they can grab him, provided he’s not a max player. There’s also a rule (#10) that prevents the same team from repeatedly winning “mix bid” auctions over a multi-year period.

Share

The auction process

3. During the auction, teams take turns nominating players in inverse order of finish. They must bid at least ⏜1 on any player they nominate. A team may pass on its nomination and still bid on players nominated by other teams, but they can’t nominate players once their turn comes up again in the rotation once they pass.

4. When a player is nominated, all teams have 7 minutes to submit simultaneous bids (reduced to 5 minutes on the second day of the auction8). All bids are then revealed, and the player is awarded to the highest bidder.

5. The maximum bid is ⏜100.

6. If 2 or more teams tie for the highest bid, the winning team is chosen by lottery among the tied bidders.

7. Teams may acquire a maximum of 4 players through ARC in any given season.

8. ARC budgets and player trades are frozen 72 hours before the auction. A team cannot select a player and immediately trade him, though in line with the NBA’s current rules, he can be traded 30 days after he signs a contract.

9. The draft ends when all remaining teams have passed on their nominations, have acquired 4 players, or have exhausted their ARC budget. There is no more “second round” under ARC, although teams may sign rookies who aren’t selected through the system as free agents.

The most important rules are #5 and #6, but they’re the ones we’ve already covered. By design, the top three picks are usually worth more than the maximum bid of ⏜100; the average #1 overall pick has a value of about ⏜140, for instance. So in most halfway decent draft classes — maybe not the Zaccharie Risacher draft — there will be multiple max bids for the first few players. Teams will face challenging decisions as to whether to acquire enough ARC to make a max bid; what’s interesting is that for every additional team that plans to bid the max, the odds of winding up with the player decline.9

I’m less hung up on some of these other rules. There is probably some slight tactical advantage to having the right to nominate a player, but it likely isn’t much.10 We might see some unconventional strategies, too, like a team nominating the equivalent of a second-round pick and hoping to get him for a few ARC while holding their fire for a bigger bid down the line; there’s nothing that says the best players have to be nominated first.11 The action would be frenetic on auction night, with new information revealed at every turn. So frenetic, in fact, that as per Rule #8, I think it’s just too chaotic to allow teams to acquire ARC on draft night. Instead, front offices would lock in their budgets and have three days to make their best-laid plans.

Multi-year planning

10. A team may win a player with a maximum ⏜100 bid at most 2 times in any 3-year window and 3 times in any 5-year window.12

11. Teams must spend at least ⏜25 in any 3-year window. It’s use it, or lose it. If a team doesn’t meet this requirement, any outstanding ARC is deducted from its account.13

12. Teams can go into ARC debt, provided they’re guaranteed to meet the ⏜25 spend per three years requirement under Rule #11.

13. Unused ARC rolls over to the next season, but with an escalating tax schedule. Teams pay a 10 percent tax rate on the first ⏜10 of ARC savings, a 20 percent tax rate on the next ⏜10 and so forth.14

14. All ARC subtracted via fines or taxes will be reallocated evenly to other teams.15 Thus, the total amount of ARC in the system is preserved.

The rollover provision under Rule #13 is intended more for “spare change” than for teams to stash large amounts of ARC. Joseph’s work for PRISM found that there’s meaningful and somewhat predictable variation in the strength of different draft classes. Still, the taxes get punitive if you try to retain more than about ⏜20 or ⏜30. And even if you can’t find a good fit for your ARC in the auction, you can always trade it.

Rules #11 and #12 are probably the more important constraints. Under Rule #1, each team is guaranteed to pick up at least ⏜25 every season. Under Rule #11, they can borrow against this and go into ARC debt, but they’re still required under Rule #11 to have enough left over to spend at least ⏜25 in any given three-year span.

On the surface, this is more flexible than the current Stepien Rule, which requires teams to make a first-round pick every other season. Under Rules #11 and #12, a team has to make the equivalent of a late-first-round pick only once every three seasons instead.16 However, under ARC, a team won’t really be able to trade draft capital more than two years in advance17, whereas draft picks can be traded up to seven years in advance under current rules. And it’s the picks in the out years that tend to be most valuable, since acquirers hope that the contending club will play through its competitive cycle and crash out to the lottery by the end of the window.

This is a big change. A lot of NBA tanking discourse takes what you might call a “demand-side approach”, i.e. by focusing on the fact that teams have a lot of desire to acquire high future picks to the point where they frequently tank. What can be neglected is the supply side: there’s a lot of opportunity to acquire such picks because teams can trade draft capital seven years out and they often take an incredibly short-term focus. Reducing the potential for long-term, franchise-compromising moves would protect contending teams (like the Phoenix Suns when they acquired Kevin Durant) from themselves. But it would also decrease the amount of draft capital available to rebuilding teams from trading their own star players. With less draft capital available, they might have to “settle” for acquiring player talent in return, which would keep them more competitive. Or they might be more inclined to keep their own players since there won’t be as many too-good-to-refuse offers.18

Rule #10, meanwhile, is intended to slightly reduce the amount of luck under our system. A team can acquire a ⏜100 player — basically, a guy who is a legitimate top 3 pick — at most two times in any three seasons or three times in any five seasons. If you’re going to spread the wealth around to losing teams, you might as well spread it a bit more evenly.19

Anti-tanking measures

15. If a team has missed the playoffs and play-in game for more than 3 consecutive years, it’s subject to a 20 percent tax on its annual ARC allocation.

16. The league will deduct ARC from teams for tanking or failing to uphold the competitive integrity of the league. If the league suspects a high likelihood of tanking, it will publicly issue a team a warning that it’s been placed under heightened scrutiny. The maximum allowable fine without a warning is ⏜5. Once a team is under heightened scrutiny, the maximum allowable fine is ⏜20 for first-time offenders and unlimited for repeat offenses, with the clock resetting after seven seasons.

Two-thirds of the 30 NBA teams make either the playoffs or the play-in tournament every season. It just isn’t that hard to do. At the end of the 2025-26 season, the only ongoing three year play-in/playoff droughts will belong to the Brooklyn Nets, Utah Jazz and Washington Wizards.

So I have no trouble punishing these teams under Rule #15. More than three years in the wilderness is when season-ticket holders start to give up hope. Would Jaren Jackson Jr., Trae Young or Anthony Davis have been shut down if the Jazz or Wizards could reset their clocks by sneaking into the play-in game? Maybe not. Under this rule, a team with a 3-year drought would actually acquire more ARC if it made the play-in than if it finished with the 5th-worst record.

Rule #16 is likely to be more controversial because it’s more subjective. But the league makes subjective judgments all the time, such as when disciplining players. It’s also fined teams like the Jazz and the Mavericks before for tanking, although the amount of the fines is trivial for billionaire owners. Rule #16 would set up a two-tiered system so teams weren’t totally blindsided; taking away a few ARC and putting them under heightened scrutiny (so for instance, they’d need to see the receipts for purported injuries) would be the warning shot. Repeated, blatant tanking once teams were on probation would be punished more harshly.

Every half-sentient NBA fan (and absolutely every person betting on the games) knows exactly which teams are tanking; it can’t be that hard for the league to police this explicitly, whether the determinations are made through a Potter Stewart standard (“I know it when I see it”) or some sort of algorithm. All you really want is a deterrent: the expected value of making dubious, tank-tastic moves will be lessened if you might face real consequences for doing so.20

Player contracts

17. All players acquired through the ARC system must be offered guaranteed contracts, with salaries tied formulaically to the winning bid and the number of maximum bids.21

18. Teams retain NBA rights to players for 4 seasons once they’re chosen through the ARC process. The ARC spent on these players is deducted from a team’s account when the player is selected, whether or not the player ultimately signs a contract. Players may decline the initial contract and negotiate a higher salary after one year, subject to certain constraints.22 But they are not eligible to sign with teams who don’t have their ARC rights until the 4-year window expires. The rights to such players may be traded, however.

We’re getting into some of the less important mechanics, so I’ll be quicker here. A player selected during the auction must be offered a guaranteed contract. Currently, first-round picks get guaranteed contracts and second-round picks don’t, but about 20 of the 30 second-round picks wind up getting full-fledged NBA deals anyway. That implies that about 50 players will typically be selected in any given ARC auction. Teams can sign unselected players later, just like they can offer contracts to undrafted free agents now.

However, under current rules, teams sometimes do something else with their late picks: choose a “draft-and-stash” international player in the hopes of bringing him into the NBA after he’s picked up a couple more years of seasoning abroad. Rule #18 is intended as a replacement for that process, while also giving international players some leverage to negotiate for a higher rookie salary if their development goes well. It also plays a more subtle role in creating another sort of “asset class” that teams can move. Since there are no more second-round picks under ARC — essentially, the spare pennies under the current system — teams could trade the rights to draft-and-stash players if they want to “round up” their capital to have the right goods to make a trade.

Transition to the new system

19. Before the start of the first season under the ARC process, each team gets an additional ⏜10 above and beyond their standings-based allocation under Rule #1.

20. Expansion teams start with 100 ARC.

21. Teams may trade ARC in advance of the full transition to ARC provided they meet Rules #10 and #11. Alternatively, by mutual agreement with the Players’ Association, the league may seek to accelerate the transition to ARC by “cashing out” existing draft-pick trade obligations to their equivalent ARC values.

Rule #19 creates a bit more liquidity in the system, but I don’t think you need or want to have too much liquidity.23 Teams can already go into ARC debt, and there’s already going to be a lot of demand for winning teams to move their ⏜25 or ⏜30 allocation to losing teams so they’ll have the capital to make a max bid.

Expansion to Las Vegas and Seattle is probable but not quite inevitable. If you’re going to ask for $8 billion (!) for a franchise in mid-tier markets, I say go ahead and give them a full 100 ARC for their first auction.

And yes, one big critique you might make of ARC is that it’s hard to have a seamless transition when teams have already traded draft picks as far out as 2032 (!). But the current system isn’t working. The alternatives proposed by the league are uninspired and might create as many problems as they solve. It’s time to rethink things from the ground up.

Most draft models try to predict how good each player will be — some number representing future WAR, or a tier classification, or an over/under on career minutes. PRISM doesn’t quite do that. Instead, it asks a different question based on pairwise comparisons: given two prospects, which one will have the better NBA career?

This design choice matters for a few reasons. Gradient boosting regression to a target like WAR has a few limitations, one of them being that the projections tend to compress towards the training set. While pairwise rankings don’t necessarily change accuracy, they preserve the structure of a gradient boosting model while allowing us more interpretability into prospect volatility, as we’ll demonstrate later. They can also reward exceptional prospects that outperform the training set.

More technically, PRISM is a machine learning model: a CatBoost gradient boosted tree classifier. Each training example is the difference between two prospects’ feature vectors. Every prospect in a draft pool is compared head-to-head against every other, producing an N×N matrix of win probabilities. The PRISM score is a player’s average win rate across all of those matchups.

Stage 1: Role prediction

Basketball evaluation is inherently position-dependent. A high post-up rate is expected for a big but unusual for a guard. A 40 percent three-point clip means something different for a point guard than for a center. Before the main ranking model can evaluate prospects, it needs to know what role each one is likely to play.

PRISM’s role predictions assign each prospect a probability distribution across three offensive archetypes (creator, spacer, big) and three defensive archetypes (perimeter, help, anchor). A prospect with 45 percent creator, 35 percent spacer, and 20 percent big is meaningfully different from one at 90/5/5 — the first is a “tweener”, the second fits the archetype. Either way, the classification is potentially useful.

These role probabilities get used in various ways by the model. They enter the ranking model directly as features. They define positional baselines, so that a prospect’s playstyle can be evaluated relative to what’s normal for their predicted role. And they drive a “tweener score”, which measures how cleanly a player maps to a single archetype.

Playstyle versatility is often an indicator of tweener status — prospects who spread their possessions across isolation, spot-up, transition, and self-creation without a dominant mode tend to have less concentration in any one archetype. But tweener-ness isn’t about current production: tweeners and clear-role prospects have nearly identical impact metrics on average in college. Whether that ambiguity helps or hurts depends on the rest of a player’s profile. Among on-ball creators and bigs, clear roles predict stronger NBA development. The league still needs point guards who run offenses and centers who anchor defenses. But among spacers, tweeners can actually develop better — the modern NBA rewards shooters who can do a little of everything without being locked into one mode.

Stage 2: Preseason score

PRISM evaluates each prospect’s current-season college statistics. But for returning players — sophomores, juniors, seniors — there’s an additional signal available: how they performed last year. A player who was already producing at a high level as a sophomore and then improves as a junior is a fundamentally different bet than a junior who’s having a breakout season after two mediocre years.

Stage 3: Pairwise ranking

PRISM then takes the outputs of Stages 1 and 2 alongside dozens of features covering box score production, advanced impact metrics, physical measurements, shot creation data, playtype frequencies, and strength of schedule, and uses them to produce head-to-head comparisons for every pair in the draft pool.

Our training data spans the 2010 through 2021 draft classes, covering the BartTorvik era where comprehensive play-by-play college statistics became available. Rather than detailing every feature, it’s more useful to describe the assumptions behind them.

Bayesian padding

College statistics are built on small samples. A prospect who shoots 15-for-40 from three has a 37.5 percent clip, but that’s 40 attempts — roughly the same number an NBA player takes in two weeks. Raw percentages at that volume are dominated by noise. PRISM addresses this through Bayesian padding, which shrinks each player’s shooting and rate statistics toward a prior expectation based on two levels of information. Every player is pulled toward a weighted blend of role-group averages — a prospect who is 60 percent creator and 40 percent spacer gets a blend of both groups' shooting expectations.

For returning players, this role-based prior is further refined by their own prior-season stats: a sophomore who shot 30 percent on 200 attempts last year will have a prior anchored mostly to his own history, while one with only 20 prior attempts will stay closer to the role group. Freshmen, with no individual history, get the role mean directly. How much a player is pulled toward the prior depends on sample size: a player with 200 three-point attempts barely moves, while a player with 30 attempts moves substantially. This padding is applied to our box score and play-by-play stats. Importantly, it happens before any composite features are computed, so downstream metrics like expected points per 100 possessions from each shooting zone are built on stabilized inputs rather than raw noise.

Teammate context and player development

Team context affects development trajectory. PRISM incorporates two features designed to separate genuine talent from circumstantial production. The first is BPM share, which measures a player’s individual BPM relative to his team’s minute-weighted average. A player with a BPM share well above 1.0 is producing far above his teammates — he’s the engine of his team’s performance, and his stats are less likely to be inflated by surrounding talent. A player near or below 1.0 is producing in line with or below his teammates, which doesn’t mean he’s bad, but it means the model should weigh his raw numbers with more skepticism.

The second is BPM trajectory, which tracks whether he’s been improving over the course of his college seasons. The raw slope is then shrunk toward the expected development rate for his class year (sophomore, junior, senior). The reasoning behind this is simple: sophomores are expected to improve more than seniors, so a sophomore’s modest improvement is treated as less noteworthy than the same jump from a senior. Freshmen receive no trajectory value since there’s no prior season to measure against. The feature captures something intuitive: a junior whose BPM has climbed steadily from 2.0 to 5.0 to 8.0 across three seasons is a fundamentally different prospect than one who jumped from 3.0 to 8.0 in a single year, even if their current-season numbers are identical. The first pattern suggests sustained development; the second might be a statistical outlier.

Age does the heavy lifting

Age, along with advanced impact metrics, accounts for a lot of the differentiation between prospects. In combination, these features separate players who are far apart in quality and development. This is by design: these metrics have proven to be predictive of future impact metrics. Age is particularly important as a lever. Players who are younger are assumed to have more physical and skill development left, which would allow them to eclipse their older classmates. But this also depends on how offensively slanted a prospect is. Older offensive prospects are assumed to be closer to their offensive peak than younger players in their class, while that same premium doesn’t necessarily exist on defense. Even older defensive centers, for example, are considered to be valuable prospects, which makes sense, as defensive development tends to be less steep than offensive development early in players’ NBA careers.

Using EPM development curves, we can see that the steepest development actually happens at age 19. Obviously, this carries inherent bias (19 year olds in the NBA are drafted because they can develop that well, and therefore receive more opportunity), but consider that seniors are still behind the curve.

You’ll notice that teams often swing for very high-upside players — this is natural when you consider the NBA’s salary structure, which punishes high-commitment contracts for non-elite production. The difference in development between players who peak at +4.0 EPM and those who peak at +2.0 EPM widens over time, and elite players stay above replacement level far longer, facing a less steep decline than merely “good” players.

There’s an interesting thought experiment that might divide many in the draft community. Say Player A has a 100 percent chance of becoming a “good” player — peaking around +1.85 EPM by year 5. Player B has a 25 percent chance of becoming “elite” and a 75 percent chance of landing at “above average.” Who would you rather have?

On pure expected value, Player A looks better: +1.85 peak EPM versus an expected +0.96 for Player B. But there’s an additional consideration in a league that caps salaries at a certain level. Instead, teams are most likely considering expected surplus.

Both players are cheap for years 1 through 4 on the rookie scale. The decision point comes at year 5, when the max extension kicks in. Player A is good enough to demand a max in restricted free agency but never quite good enough to justify it. This is the Domantas Sabonis contract, the Ja Morant contract — players in the gap between +2.0 and +4.0 who get paid like stars and produce more like solid starters.

Player B profiles differently. In the 25 percent scenario where he becomes elite, you have a franchise cornerstone producing +4.7 EPM on a max deal — massive surplus, and the kind of player championships are built around. But if he’s "merely” above-average, you simply don’t (or at least you shouldn’t) extend him at the max. He’ll walk into free agency or re-sign at a discount, and you’ll lose nothing beyond the draft pick and rookie-scale investment.

This is exactly why teams often optimize for different outcomes in the NBA — combined with aging curves, it’s unlikely that older, creation-driven prospects will warrant the type of investment in the NBA that gives them positive EV on max contracts. Hence, those players tend to fall in the draft.

Of course, this makes selecting the right players at the top even more important. An important but sometimes overlooked part of development curves is how they begin — future “elite” players post +0.01 EPM as rookies, essentially neutral, while the next tier down is already at -0.93. The gap between tiers is visible from year one, even though few rookies are actually good in an absolute sense. The implication isn’t that teams should draft for immediate production — it’s that the signal is there early if you know where to look. PRISM is designed to find it: by anchoring on current college production adjusted for age and context, it can separate which young prospects are tracking toward elite trajectories versus which are merely projectable. The difference between those two outcomes, as the aging curves show, is the difference between a franchise-altering max contract and a roster-clogging one.

Custom features decide close calls

Where we hope PRISM’s edge comes from is a set of custom-engineered features that differentiate similarly-ranked prospects. In head-to-head matchups between players at similar impact levels, secondary features often decide the outcome.

One of the more predictive features, steals, has long been known as a cognition metric, suggesting an understanding of positioning, which becomes valuable for both creator archetypes and big men. That intuition is confirmed when analyzing EPM production: positive relative STL/100 players have steeper developmental increases as players continue deeper into their careers.

Position-adjusted playstyles measure how a player’s shot diet and play-type distribution compare to others in their predicted roles. A big who shoots more than the average big, or a guard who posts up more than the average guard, represents a meaningful deviation from positional norms. These deviations help the model identify stylistic outliers whose production might translate differently than their archetype would suggest.

Self-creation metrics from play-by-play measure how much of a player’s scoring comes off their own creation versus assisted looks. This is one of the clearest distinctions between scalable and role-dependent production. A player who makes 3 unassisted threes per 100 possessions is demonstrating a skill that is valued in the NBA; a player whose threes are almost entirely assisted is more context-dependent.

Still, perhaps too much emphasis is being put on self-creators in draft outcomes. The data bears this out — high self-creation in college doesn't particularly predict stronger NBA development.

Physical measurements are amplifiers, not drivers

Height, wingspan, length (wingspan minus height), and BMI are included in the model, but they function more as amplifiers on top of existing production rather than as standalone predictors. Fifteen years of data tells us that combine measurements and athletic testing are noisy as standalone signals. If you want an athletic guard because athletic guards get to the rim, a guard who already gets to the rim is a good bet, regardless of how he tests at the combine.

Position-adjusted physicals — how tall or long a player is relative to their predicted role — carry more weight than raw measurements. Being 6’9” matters differently depending on whether you’re projected as a guard or a big.

Wingspan is an interesting feature, particularly because its importance differs by position. For perimeter defenders, length (the difference between wingspan and height) matters more. For anchors, pure wingspan matters more. For helpers, neither has particular importance — instead, those players indicate value as defenders with activity, by generating stops, steals and blocks, and having high defensive impact metrics in college. Rebounding also generates positive outcomes and shouldn’t be ignored as a part of defense.

Rolling windows and temporal integrity

Prospects are scored within three-year rolling pools. The 2026 class is pooled with 2024 and 2025; the 2025 class with 2023 and 2024; and so on. This keeps PRISM scores interpretable across years by providing a consistent reference frame. The pooling approach also allows us to get around era effects. Since players aren’t going to be compared against players ten years in the past or future, we aren’t penalizing prospects from previous classes for less “modern” playstyles.

Training pair construction

Not all training comparisons are equally informative. PRISM distinguishes three types of pairs:

Survivor–Survivor: Both players earned meaningful NBA minutes. These are the most valuable comparisons and are included without modification.

Survivor–Bust: One player stuck, the other didn’t. These are downweighted. The reasoning: separating NBA players from non-NBA players is relatively easy and not the model’s primary job. We want capacity allocated to the harder problem of distinguishing between viable prospects.

Bust–Bust: Both players washed out. These comparisons are excluded entirely. The ordering between two players who never stuck in the league — did a guy provide below-replacement-level production for one season or two? — is noisy at best.

Why wins above replacement?

For our player comparisons, we use eWINS — wins created above replacement level according to EPM, setting negative seasons to zero — rather than a player’s rate of production as measured by EPM. This is among the more “controversial” design choices of the model, but we think it’s the right one. eWINS account for playing time and volume, without punishing players for below-replacement-level performance if they’re still learning on the job. The model’s primary job is ranking prospects against each other, and most of the separation between draft outcomes isn’t between good NBA players and slightly better NBA players. It’s between players who make it and players who don’t. Roughly half of all drafted prospects never establish a meaningful NBA career.

EPM can’t cleanly represent this. A player who never reaches the NBA has no EPM. You’re forced to either exclude him from training entirely or invent a penalty value and hope it lands in the right place. But you can calculate his eWINS: 0.

More specifically, we sum eWINS across a player’s first seven NBA seasons, with heavier weight on the later windows where minutes are earned rather than granted by draft hype. This means the target rewards sustained production, not just a single outstanding year. A player who is a solid starter for six seasons accumulates more eWINS than one who has two brilliant years and washes out — and that distinction reflects real draft value.

Scouting-based rankings serve as PRISM’s prior

PRISM has no access to combine athletic testing (vertical leap, agility, sprint times), interview or character intelligence, medical evaluations, or coaching assessments. This is partly a data limitation and partly by design. The model is scoped to statistical production and measurable context instead.

But to compensate for this missing information, we incorporate a prior based on consensus scouting rankings, and this is rated more highly toward the start of the season when statistical metrics of production aren’t reliable either. In early November, a prospect may have played three games. His statistics are essentially noise. But consensus draft boards — aggregated from scouts, media, and market signals — contain real information about a player’s talent level, even if imperfect. As games accumulate, the model’s evaluation gradually takes over, and by roughly early February, the model’s statistical measures dominate.

The mapping from consensus rank to expected PRISM score is not linear. The gap in expected outcome between the consensus #1 and #5 prospect is far larger than the gap between #30 and #35 — the tails are steep. We calibrate this mapping empirically using a non-parametric method that captures the shape of the tails.

We also adjust the prior to account for variation in class quality. The consensus #10 prospect in a historically loaded class represents a meaningfully different talent level than the #10 prospect in a shallow class. If we applied the same anchor to both, we would systematically undervalue prospects in strong classes and overvalue those in weak ones. To address this, we compute draft strength scalars for each class using the pairwise matrices, which compare every prospect against every other prospect in the same rolling pool. The scalar measures how often prospects from a given class beat prospects from adjacent classes in the model’s head-to-head evaluations. Lottery-range and late-class prospects receive separate scalars, since a class can be deep at the top without being strong overall, or vice versa.

PRISM vs. consensus

PRISM isn’t afraid to take big swings. In backtesting, among players who earned NBA minutes, PRISM has a significant edge over consensus rankings in pairwise accuracy. Consensus rankings, which correlate strongly with draft capital and therefore opportunity, are better at “finding” players that stick in the league. But among the players who do stick, consensus is barely better than a coin flip at telling you who will be better.

PRISM’s biggest raises have surfaced real value — typically productive college players whose translatable skills get discounted because they lack the pedigree or physical tools that scouts fixate on. Multiple players that ranked significantly above their consensus position became quality NBA contributors with positive impact metrics.

Volatility scores, trajectories and draft simulator

Using our pairwise probabilities, we can assess how a player's feature profile translates against different tiers of competition. A player's volatility score — their average win probability against the top K prospects vs against the bottom K prospects — captures how consistently they separate from the better players in their draft class. A prospect whose ranking is driven primarily by dominating lower-tier prospects, rather than competing with the top of the pool, may have a lower ceiling but a higher floor.

Additionally, we measure which players peak earlier or later. These features, along with PRISM scores and role projections, give us a good starting point for something fun: a draft simulator. Using roster composition, player labels (franchise player, core, expendable, rotation), and team designations (blank slate, rebuilding with assets, rising, contending, aging contender), we can develop a better sense of which players a team would logically target. Generally, rebuilding teams will draft the best player available, while teams on a contending timeline would prefer prospects who would make an immediate impact and fit within their current roster scheme.

PRISM can’t account for everything

PRISM is best treated as a signal, not a verdict. The model has no access to medical or injury information, interview intelligence, or coaching assessments.

Draft projection is inherently imperfect. NBA decisions and outcomes reflect more than college production alone, even if college production is the strongest indicator.

Last updated 3/15/2026. We’ve added information on the women’s version of COOPER, as well as how our NCAA tournament foreacst work. As we’ve been working on these, we’ve also made a few minor tweaks to the parameters of the men’s version. Although the differences are hard to notice, the text below reflects the current values for all parameters.

What’s new in COOPER

COOPER — named in honor of Naismith Award winner Cooper Flagg and two-time NCAA champion Cynthia Cooper — is Silver Bulletin’s new college basketball rating system.1 As part of a long Nate/Silver Bulletin tradition, it is a goofy backronym: Collegiate Outcomes with Opponents, Pace and Expert Ratings. The name hints at the system’s essential features:

• The first “O” is for “outcomes”. COOPER is mostly based on margin of victory, but simply winning games matters, too.

• The second “O” stands for “opponents”. The model adjusts for the strength of opponents through an Elo-like system, which is essential when rating college teams with their wide disparity in quality. Furthermore, games between more evenly-matched teams are weighted more heavily in the system.

• The “P” for “pace”. We now account for how high-scoring a team’s games tend to be and derive both offensive and defensive ratings for each school. Higher-scoring games tend to introduce more variability.

• Finally, “ER” for “expert ratings” means we revise each team’s rating at the start of the season using human opinions in the form of the preseason AP and Coaches’ polls. A team’s initial rating to start the season is based on a combination of these polls, its year-end rating from the previous season, and the strength of its conference.

COOPER represents an evolution of the SBCB ratings that we used in 2025, and COOPER and SBCB share some of the same programming, but we’ve dug pretty deep under the hood and there are some important changes:

• The most noticeable one is that we now calculate separate offensive and defensive ratings for each school, which we call PPPG (projected points per game) and PPAG (projected points allowed per game). Essentially, these ratings represent how many points we’d expect a team to score against an average NCAA opponent. For instance, a team with a PPPG of 81 and a PPAG of 74 would be expected to win 81–74. By subtracting PPAG from PPPG, we can also derive a net rating for each team: in this case, it would be +7.

• But net rating doesn’t tell the whole story because teams that play to higher scores (both scoring and allowing more points) tend to have higher variance. Instead, a team’s Elo rating accounts for this property and is its best representation of its likelihood of victory against an average opponent.2

• Although it’s less visible, perhaps the most important change is that we’ve removed the constraint from SBCB that required a team to always gain in the ratings when it won and always lose points when it lost. For instance, even if Duke was a 35-point favorite against Cal St. Bakersfield and won the game by 1 point at the buzzer, our previous system would have viewed this as a positive for Duke. But that’s pretty clearly an unreasonable assumption from a Bayesian standpoint — you wouldn’t think more highly of Duke after this game — and this was causing a fair amount of information loss.3 So this is no longer the case in COOPER. Teams are awarded a bonus for winning games, regardless of the final margin.4 But they can now lose ground if they significantly underachieve COOPER’s expectations even after a win — or gain ground following a loss where the final score is impressive relative to the model’s expectations.

• COOPER tends to show a wider spread between the best and worst teams than SBCB did. This is partly because there’s now slightly more carryover in the ratings from season to season. Previously, our view was that the increasing likelihood of star players defecting to the NBA after just a season or two was diminishing the advantage for blue-blood teams. However, NIL is helping the most elite programs remain as dominant as ever. Also, as described above, COOPER tends to put more emphasis on margin of victory as compared with SBCB and doesn’t punish teams for runaway margins in the same way. This is a bit less “politically correct” in the sense that it potentially gives teams more credit for beating up on weaker opponents. But it increases the fidelity in distinguishing good teams from great teams. Predictive accuracy is what we’re after here.

• However, each game now has what we call an “impact factor” that reflects how reliable a signal it provides. Basically, games that are projected to be lopsided5 matter less in COOPER than those that the model expects to be close. Conference games and especially NCAA tournament games also have higher impact ratings. Teams tend to compete at full effort in these games, and the results are more reliable than for early-season, non-conference matchups.

• Although COOPER continues to account for travel distance — an East Coast team playing in California will often have a rough time — we’ve found that these effects are diminishing over time as travel accommodations improve to a larger extent than SBCB was accounting for. However, we’ve retained one of my favorite SBCB features, which are customized home court advantage ratings for each school.

• SBCB calculated both a “Bayesian” version of the ratings that adjusted each team’s rating every offseason based on preseason polls, and a “pure” version that applied purely objective data. We’re continuing to do this, but the “pure” version now receives less emphasis. The poll-adjusted version should be considered the official or flagship6 version of COOPER.

More about how COOPER works

Just so I don’t get accused of self-plagiarism, note that some of this text is copied from the SBCB methodology page.

• COOPER is a profoundly Bayesian model in the sense that ratings are adjusted on an iterative basis as new information becomes available: namely polls at the start of the season, and then game results throughout the season. We don’t go back and revise past COOPER ratings based on information that wasn't available at the time the game was played.

• Rating changes after each game reflect a combination of:

  • The margin of victory or defeat as compared with COOPER’s prior expectations of the “point spread”. Unlike in previous versions of SBCB/Elo, there are no diminishing returns to higher scoring margins. In practice, both NBA and college basketball teams tend to lay off the gas pedal once they already have a big lead, so the differences in the final score can actually understate the intrinsic gap in team quality. A linear formula works perfectly well for predictive purposes.

  • But we also account for whether a team wins or loses. Winning a game by any margin is essentially equivalent to 6 points of scoring margin. However, this “bonus” is compared against COOPER’s prior for each team’s probability of victory. Therefore, it has little effect if a team was heavily favored because the expectation of a win was already baked into the system. Wins in closely contested matchups — or upsets in games with clear favorites — matter far more.

• Home court advantage is factored in. In fact, we calculate a separate home court rating for each team, based on how much it underperforms or exceeds its COOPER projection in home games. Generally speaking, teams that are reputed to have larger home court advantages based on difficult playing conditions or more enthusiastic fan bases actually do. Teams that play at high altitudes often have especially large home court advantages in basketball, as in other sports. These home court ratings move very slowly, taking advantage of data from previous years. (They fully carry over from season to season.) However, having a larger home court advantage isn’t helpful in the NCAA tournament, which is entirely played at neutral sites. Teams like Purdue, whose home court is worth an additional ~2 points of victory margin compared with the NCAA D1 average, may be overrated by other systems that don’t account for this factor.

• Travel distance also matters and, for the 2025-26 season, is equal to 5 * m^(⅓) worth of Elo ratings points, where m is the distance in miles from the visiting team’s campus. For home games, be sure to add the travel distance factor to the team-specific home court rating to calculate a team’s overall advantage. But note this is a considerably smaller advantage from travel than SBCB had assumed. We found that SBCB had been overrating home teams in recent years in games where the opponents traveled a long way to play. On the other hand, the effect of travel distance was much larger up through the 1980s. This almost certainly reflects improving travel accommodations and sports medicine and the general professionalization of college sports.

• COOPER ratings, like most other Elo systems applied to sports, partly carry over from season to season, with a discount factor applied that reverts the ratings toward the mean in between seasons. To be absurdly specific, a team’s rating is reverted by 30 percent toward the mean at the start of each new season. This is actually less mean-reversion than had been incorporated into SBCB. While it’s true that elite college talent tends to go pro sooner, top-tier programs like Duke or Kansas have plenty of other ways to perpetuate their success by investing more in their programs or through superior recruiting. While the introduction of NIL several years ago was “disruptive” to some degree — for instance, in boosting the basketball profile of the SEC — recent tournament and regular-season results suggest a recalibration toward a new steady state.

• More precisely, teams in COOPER are reverted toward the mean of other teams in their conference — not toward the global average (with the exception of the few remaining independent teams). When a team changes conferences, the rating change is based on its new conference rather than its old one, as this can indicate where a program fits into the college basketball pecking order. Interconference play, especially in recent NCAA tournaments, is self-evidently important for this purpose. In essence, a team that exceeds expectations in the NCAA Tournament will then redistribute those gains toward the rest of its conference in COOPER’s off-season recalibration process. The default/Bayesian version of COOPER also accounts for preseason polls in its initial ratings to start the season, while the “pure” version does not.

• However, this introduces various complications because the polls only provide a truncated list: that is, only 25 teams are ranked.7 The process for imputing human ratings quite literally applies Bayes’ Theorem in the sense that it relies on a prior about how likely a team is to be ranked. For instance, a team with a 2000 Elo rating would typically expect to be ranked somewhere in the top 25 in the next preseason poll — so if it isn’t ranked, that provides a lot of information that its performance is expected to decline, usually because of a loss of key talent. However, a team that ended the previous season with a mediocre rating would rarely expect to be ranked, so this tells us little. Teams ranked specifically #1 overall receive special treatment to ensure that truly dominant clubs like the late 60s/early 70s UCLA Bruins are not punished.²

• I haven’t yet mentioned what is perhaps the most important parameter in any Elo-derived system, which is called the k-factor. This governs how much the ratings update after each game. A higher k-factor implies more sensitivity to recent play but also more volatility. Statistically speaking, the goal is generally to minimize autocorrelation. That is, you want to avoid both a too-high k-factor where ratings zig-zag around (i.e. teams usually decline after gaining and vice versa) and a too-low one where a team with a recent ratings gain can predictably be expected to follow that up with further gains because the system is too slow to account for what soccer fans call a change in “form”. Specifically, we use a k-factor of 55; this number has no intrinsic meaning and is derived empirically. Generally speaking, COOPER ratings are more aggressive than other college basketball systems about accounting for recent play and tend to ride a winning hand while discounting ratings for teams that have been on a downward trajectory.8

• However, the k-factor is up to 2x higher (so, it goes up to 110) for early-season games, diminishing until a team plays roughly the 15th game of its season. The intuition behind this is that early-season games reveal a lot of information as compared to COOPER’s crude preseason estimates. By the middle of the season, conversely, teams mostly are “what they thought we were” and each subsequent game tells us less.

• Newly this year, each game receives an impact factor. Basically, games between teams that are closely matched in COOPER matter more. We’ve also found that what we call “high-stakes” games — namely, conference games and NCAA tournament games — tend to provide more signal, and these are weighted more heavily also. For NCAA tournament games specifically, there’s also a hard-coded boost to the impact rating. We’ve found that success early in the NCAA tournament — i.e. if a team blows out tough opponents in the first two rounds — tends to predict success for the rest of the tournament.

• For calculating margins of victory and net ratings, one point in a basketball game equals approximately 28.5 Elo points. Thus, a team with a 100-point Elo advantage, after accounting for home court and travel distance, would be roughly a 3 or 4-point favorite. However, newly for COOPER, this exchange rate varies slightly based on whether COOPER projects the game to be high-scoring or low-scoring.

• Also new this year, COOPER calculates a rolling “pace” factor for each team. This is a slight misnomer, because in basketball analytics, “pace” generally refers to the number of possessions in a game. For COOPER, because possession-by-possession data is unavailable until recent seasons, it instead reflects the overall number of points scored by both schools in games involving the team. In addition, we calculate a Bayesian expectation of the NCAA average points per game: 2025-26 has been a particularly high-scoring season, for instance. This leaguewide rating is designed to adjust more aggressively at the start of each season; changes in rules or style are often evident relatively quickly.

• The combination of a team’s net rating and its pace rating essentially allows us to back into an offensive and defensive rating for each school.9 Higher-scoring games tend to introduce more variability. Both empirically and theoretically, a marginal point matters more in an environment where points are scarce. For instance, an uptempo team that projects to beat an average opponent by a score of 90-80 will win less often than a downtempo team that projects to win 65-55. Conversely, teams with offense-oriented mindsets tend to pull off a few more upsets when they’re underdogs, such as by winning the battle for 3-point shooting.

• Teams that are new to Division I begin with a rating of 1300 at the start of their first D1 season, adjusted slightly upward or downward based on the strength of their new conference. That is to say, they are usually considerably below average since the average Elo rating is 1500. Preexisting D1 teams’ ratings are adjusted slightly upward such that the global average remains at 1500 when new teams join.

• Our database contains many games between Division I and Division II teams, especially in recent seasons. However, rather than calculating a rating for individual D2 teams, we instead lump all D2 teams together into a single “divtwo” rating. Essentially, this makes them the equivalent of the Washington Generals, barnstorming around and usually getting obliterated.10 A separate “divtwohome” running tally is calculated for D2 teams that host home games as opposed to paying on the road or at neutral sites — but this has become rare as D1 teams generally don’t want to decline an opportunity to sell tickets. Overall, however, D2 teams are patsies, with D1 teams winning upwards of 99 percent of the time at home in recent years against D2 opponents.

• COOPER and our NCAA tournament model estimate win probabilities for each game by forecasting a game score and estimating a standard error of its forecast. The standard error is higher in lopsided matchups and in “low stakes” games (i.e., nonconference matchups outside of the NCAA tournament). The statistical distribution we use is slightly “fat-tailed”, meaning that outlier scores occur somewhat more often than one might predict from a normal distribution, possibly indicating unusual circumstances (e.g., the entire road team gets food poisoning). (Added 3/14/2026).

Differences in women’s COOPER

The women’s version of COOPER essentially uses the same code as the men, but with a few parameters tweaked to reflect the nuances of the women’s game.

• There tends to be more dominance among women, i.e. the top teams win more often against average ones, and by larger margins. This mostly emerges organically from the rating-generation process, although there are a few small changes. Division II teams and new Division I teams start with lower default ratings than in the men’s version, and there is a considerably wider spread introduced in the Bayesian version of the ratings.

• There is less mean-reversion from season to season, probably because women are not eligible to join the WNBA until age 22, leading to greater team continuity. Thus, 80 percent of ratings carry over from season to season instead of 70 percent for the men.

• Home court advantage and travel distance effect tend to be slightly less in the women’s game. The top women’s programs draw crowds in line with men’s teams, but obscure schools may literally play in gyms in front of a few hundred fans. As for the men, home court advantage is calculated on a team-by-team basis. Note that Round of 64 and Round of 32 NCAA tournament games are often played on home court sites in the women’s tournament.

• The conversion rate between Elo points and game scores is slightly different for women. One point in a basketball game equals 25 Elo points for women, as compared 28.5 points for men. This likely reflects the fact that women’s games are typically more lopsided, so a team gets slightly less credit for running up the score. As for men, these are default values that are adjusted based on a team’s pace.

• Women’s ratings are based on the 2002-03 season onward rather than 1949-50 for the men. Data is also somewhat less complete: for instance, more games against D2 opponents are missing, and data on game locations for neutral-site games is much less comprehensive.

NCAA Tournament forecasts

After it spent many years living in an exceptionally complicated EXCEL spreadsheet that still contained some code I wrote when trying to win my office pool back in ~2002, I’ve finally transitioned our tournament forecasts into some actual code.

But our NCAA tournament forecasts aren’t really all that complicated. They account for differences in team strength, game location, and on the women’s side — since top seeds usually host games in the first two rounds — home court advantage. COOPER already calculates a probabilistic forecast for each game as part of its ratings, so basically, we’re just applying that process going forward to the most important 67 games of the season.

There are, however, three important differences to plain old COOPER. One is that the simulations run “hot”, meaning that we account for the effect on COOPER ratings from simulated games in earlier rounds. Conditional upon winning, a team’s rating is likely to improve, i.e. if a #12 seed defeats a #5 seed, the 12-seed is probably better than we originally expected, and that should be accounted for in forthcoming games. However, this effect is stronger for underdogs because they’re not expected to win, meaning that they require a larger Bayesian update. If and when a 12-seed reaches the Final Four, it was probably the case that they were underseeded to begin with. Thus, the late rounds of the tournament can actually be less chalky than the earlier ones.

The next difference is that our tournament forecasts account for injuries, whereas regular COOPER does not. The injury information is probabilistic; we have to make some judgment calls on translating sometimes subjective injury reports into probabilities that the player is sidelined for forthcoming rounds. For instance, if we list a player as having an 80 percent chance of playing, he’ll appear in the game 80 percent of the time and be sidelined the other 20 percent. Once a player is simulated as returning to his team’s lineup, he’ll remain with the team in that simulation for all subsequent rounds.11 (We don’t attempt to forecast new injuries.)

The model also accounts for significant players who missed time during the regular season but are now available. Thus, some teams will have an upward injury adjustment for having gotten healthier.

For what it’s worth, we also put a fair amount of time into double-checking the injury logic this year, including reevaluating the magnitude of the effect they had our projections, and I think our previous models were probably understating their impact. Top college players, like the top NBA players, can impact the final margin by as much as 7-10 points versus a replacement level player12

Historically, data on women’s injuries is sparser and we haven’t accouxnted for it, but we’re including whatever information on women’s injuries we can track down.

The injury adjustment accounts for the importance of the player, as measured by sports-reference.com Win Shares, adjusted for strength of schedule. We project a player’s replacement based on an historical analysis of players who played 5 MPG or less, adjusted for team strength. In other words, even a scrub on Duke who fills in for an injured player is likely to be pretty good, whereas one on Southeast Missouri State might not be. Thus, the most devastating injuries are when a middling team loses one of its few stars.

Finally, our tournament ratings combine COOPER with other leading systems, namely the Pomeroy ratings for men and Her Hoop Stats for women. In our tournament forecasts, COOPER gets 5/8ths of the weight and the other systems get 3/8ths.

Previously, we’d used a composite consisting of several other rating systems, so paring down to just Pomeroy and Her Hoop Stats reflects a simplification. However, there are several reasons why we prefer this pared-down approach:

• Having spent a lot of time looking at different ratings systems, I believe that these are the most rigorous ones (alongside COOPER). They also tend to be highly correlated with COOPER.

• Pomeroy and Her Hoop Stats calculate offensive and defensive ratings in addition to overall ratings, which makes them more directly comparable to COOPER because COOPER does that do. Most power ratings do not do this.

• Reducing the number of third-party systems reduces the workload on Selection Sunday and allows us to turn around the tournament forecasts more quickly, while reducing the probability for errors.

Pomeroy and Her Hoop Stats ratings are adjusted to give them the same mean and standard deviation as COOPER. They’re also adjusted for team pace. The overall ratings listed by Pomeroy and Her Hoop Stats are actually efficiency ratings, i.e. a measure of how many points a team is expected to outscore its opponents per 100 possessions. However, because some teams play at a faster pace than others, they’ll average a different number of expected possessions in a game. A team that is expected to outscore its opponents by 0.2 points per possession but averages 70 offensive possessions per game might be more efficient than one that averages +0.19 points per possession but plays at a pace of 80 possessions/game, but the latter team will actually outscore its opponents by a wider margin than average because it has more opportunities to press its advantage. This accounts for some of the difference between Duke and Michigan this year in different ratings systems; Duke is slightly more efficient on a per-possession basis according to most systems, but Michigan plays at a considerably faster pace.

Like in COOPER, the rating composite used in our model adjusts after every game using an Elo-like formula. In fact, the formula is identical to the COOPER formula, other than that our tournament version also accounts for injury status at the time of the game. An outcome that is out of line with the model’s expectations — a big underdog wins outright, or a 10-point favorite wins by 30 points — will cascade into a team’s rating for the rest of the tournament. Note also that tournament games have a larger impact factor in COOPER than regular-season games, so the early rounds of the tourney can potentially have a big impact.

We’ll update this document if we catch any bugs or make any further changes. Thank you for being a subscriber to Silver Bulletin and for your interest in COOPER!

Subscribe now

A year ago, after a trade so shocking that it literally made me wonder whether I was hallucinating1 sent Luka Doncic to Los Angeles, I ranked all 30 NBA teams by a deceptively simple criterion: their likelihood of winning future championships over the next 10 NBA seasons. The Future of the Franchise rankings have since turned into one of my very favorite Silver Bulletin features. In fact, we repeated the exercise in July, inviting two additional judges, Jeremias Engelmann and Joseph George, to the panel. (Joseph is Silver Bulletin’s Assistant Sports Analyst; Jeremias is the developer of Real Plus-Minus and a former analyst for the Mavericks and Suns whose excellent work can be found at xRAPM.com and the highly recommended 5x5.)

The timing is perfect for another round of FotF. The NBA will return to action tonight after an All-Star break that featured more discussion about tanking than basketball. To be honest, we’ll probably invite ourselves to the tanking party at some point. (There’s an anti-tanking take halfway done in the drafts folder.) But today, let’s talk some hoops.

For most of the league — there are roughly a half-dozen exceptions — we can find at least something to feel optimistic about. But let’s get this out of the way: we’re unapologetically ringmaxxing here. Consistent with NBA’s ring culture taken to its logical extreme, titles are literally all that matter for FotF, not mere pretty-goodness.

More specifically, we’re looking at the span from this year’s championship (2026) all the way through … 2035. I used the plural of “championship(s)” because multiple titles count more.2 Although there have been seven distinct title-holders in the past seven NBA seasons, that’s, historically speaking, an outlier. We’re looking at which teams have the ingredients to build a potential dynasty.

The talent on the roster matters, obviously — especially high-end talent. But so does draft equity, the cap sheet, managerial acumen, the allure of each market in attracting future talent in the player empowerment era — and the timelines for getting all of these things to perfectly align when the default chance of winning a championship is just 1 in 30. Indeed, I don’t think there’s any strictly algorithmic way to do this3, and there’s quite a bit more disagreement among the panel than during the summer, especially on teams like the Pacers and the Hawks. You’ll see a few consistent themes in these differences: I (Nate) tend to weigh franchise prestige more heavily than Jeremias or Joseph; Jeremias puts more of a premium on management, and Joseph is very high on this year’s draft and so is more forgiving to teams that are tanking at all costs.

Teams are ordered by the average of our rankings, with ties broken by the median. The arrows indicate changes from the July version, with thicker arrows (⇧ or ⇩) signaling moves of at least 5 positions. The fancy charts we tried last time didn’t quite work, so we're going for a more retro look this time, emojis and all. (Yes, we’re aware that a 🦩 is a flamingo, not a pelican; please register your complaints with the Unicode Consortium.) In line with FotF tradition, we’ll start from the bottom of the league and count up. That means we’re going to force you to endure a few paragraphs about the Sacramento Kings before we get to the good stuff; the biggest disputes usually tend to be about teams that rank somewhere between #15 and #7 on our list.

30. 👑 Sacramento Kings ↔

(Nate: 30 ↓, Joseph: 27 ↑, Jeremias: 30 ↔)

Jeremias: The Kings remain a laughingstock — they’ve appeared in the playoffs once in 20 years. This year’s annual attempt to make the postseason landed so short that they essentially fell backwards into doing what it takes to get better: losing for a better draft pick. In fact, the Kings have the NBA’s worst record and best lottery odds.

I’d give them more credit if this were planned, but it’s basically a function of two things: a lousy roster and an injury to Domantas Sabonis. Given the owner’s rather ham-handed use of analytics and questionable decisions about who to put in power, the Kings will probably continue down the path of NBA irrelevancy.

Joseph: I’m slightly more optimistic on the Kings than consensus. Yes, they’re the worst team in the NBA — but that’s precisely the point. As long as they keep tanking, the floor is the fifth pick, and this draft isn’t the big three that mainstream outlets keep pushing. It’s more like a big five. (Our NBA draft player ranking model, PRISM, will launch soon — Nate.) The drop-off in expected value doesn’t hit until pick six or seven, which means Sacramento is in line for a great prospect regardless of lottery luck.

That alone should put them ahead of New Orleans, who doesn’t even own their pick, and Chicago, who seems hell-bent on stacking guards and making the play-in tournament every year.

29. 🦩 New Orleans Pelicans ↓

(Nate: 29 ↔, Joseph: 30 ↓, Jeremias: 27 ↑)

Jeremias: The Pelicans caught a lot of flak for their 2025 draft-day deal, giving away a future “superfirst” — the best of their own and Milwaukee’s picks in 2026 — to Atlanta in exchange for Derik Queen. While Queen’s rookie season has given rise to cautious optimism, he still doesn’t represent nearly the expected value of the pick the Hawks are about to receive, which lands in the top three in the NBA lottery in 50 percent of my simulations. And this is an especially talented and top-heavy draft class.

But this predicament is just a symptom of a bigger, underlying problem: the grandiose confidence and ineptitude of Pelicans management.

And with Zion Williamson’s impact fading and rookie Jeremiah Fears shaping up as one of the NBA’s worst defenders, the Pelicans have no true superstar talent to build around. Add it up, and it’s hard to imagine New Orleans winning a playoff series in the foreseeable future, much less a championship.

28. 🧙 Washington Wizards ↓

(Nate: 25 ↓, Joseph: 28 ↔, Jeremias: 28 ↓)

Nate: It was, I think, at a Super Bowl party where I encountered a Wizards fan who was excited about the Trae Young/Anthony Davis combination. It does improve the floor, but FotF is explicitly about the championship upside, and both Young and Davis are probably negative-value contracts.

It doesn’t negate the Wizards’ other assets — Alex Sarr is eventually going to be a solid rotation player, and possibly well above that. And in addition to Sarr, you’ve got … well, I’m not sure. This is typically the point at which I’d highlight an unheralded asset. But apart from Trae and AD, the Wizards don’t currently have any regular player with a positive EPM. OK, let’s round up on this one: Tre Johnson has an exactly league-average EPM (±0.0), and is shooting 38.6 percent from 3. That’s promising, offsetting perceptions that he was one of the more likely lottery busts. Still, I’m not sure there’s a plan here. Speaking of which…

27. 🐂 Chicago Bulls ↓

(Nate: 26 ↓, Joseph: 29 ⇩, Jeremias: 26 ↓)

Jeremias: I agree with Bulls GM Arturas Karnisovas, who said earlier this month, “It’s being in the middle. That is what we don’t want to do.” But the Bulls’ moves have not indicated much deviation from their multiyear approach of chasing a low playoff seed. When it comes to winning titles, the stated goal in Chicago, that’s probably the worst strategy to follow.

The Bulls did something peculiar at the deadline, trading for three of the worst defenders of the past three decades, according to 30-year Adjusted Plus-Minus.

Perhaps bad defense is part of the plan to become a truly bad team. But these players probably aren’t quite bad enough, in total impact, to sink the Bulls sufficiently in the standings. So the treadmill continues.

26. 🦌 Milwaukee Bucks ↔

(Nate: 27 ↓, Joseph: 23 ↑, Jeremias: 29 ↓)

Jeremias: Since the Bucks’ 2020 trade for Jrue Holiday — a successful move that led to an NBA title — their strategy has been to go all-in over and over again, mortgaging their future with win-now moves designed to placate Giannis Antetokounmpo.

But now it’s time to pay the fiddler. The Bucks are no longer even a likely play-in team, with the roster beyond Antetokounmpo sporting very few positive-impact players.

Giannis seems to perpetually have one foot out the door, with many expecting the Bucks to finally trade him this offseason. But even if he stays, that doesn’t augur an especially bright future, as he’s already 31 and now rather injury-prone. At least a Giannis trade would mitigate what appears to be a brewing disaster: The Bucks look sure to bottom out, and yet their high draft picks will keep going elsewhere, as they don’t control their first-rounders from 2027 to 2030.

Nate: For what it’s worth, I think the decision not to trade Giannis at the deadline was logical. Cap sheets and 2033 draft picks will free up in the summer. And currently-injured players present an asymmetric information problem that makes realizing fair value harder. But there’s been a paradigm shift around the league under the new-ish de facto hard cap toward recognizing that veteran max players with injury histories and/or obvious “fit” issues very much present franchise-crippling downside risk (as well as upside). You saw that in the low price for KD last summer. The apparent disinterest in Giannis from the likes if the Spurs, Rockets, Thunder4, Hawks, et. al., suggests the Bucks might suffer from a lesser version of the same phenomenon.

25. ⛵ Los Angeles Clippers ↔

(Nate: 28 ↓, Joseph: 24 ↑, Jeremias: 24 ↓)

Read more

You can find ELWAY’s Super Bowl odds at the ELWAY landing page here. But today’s newsletter provides a lot more detail about the game. To be honest, today’s newsletter is a bit overstuffed — rather than hitting your inbox two or three times amid a busy political news cycle, I’m sending one comprehensive Super Bowl preview considering:

• Where the Seahawks and Patriots would rank among the most unlikely Super Bowl champions;

• ELWAY’s retrospective ratings for every past Super Bowl;

• QBERT ratings for every past Super Bowl quarterback;

• A deeper look at the Patriots’ schedule;

• Is Sam Darnold too mediocre to win the Super Bowl? And is Drake Maye too young or too injured?

• Why Super Bowls tend to be high-scoring — just in case you’re thinking about betting the over/under line;

• And, the most likely exact final scores and the best Super Bowl squares based on 30,000 ELWAY simulations, including a downloadable EXCEL spreadsheet.

The Patriots and Seahawks previously faced off in Super Bowl XLIX in Glendale, Arizona. That game ended spectacularly when Russell Wilson was picked off by the Patriots’ Malcolm Butler on 2nd-and-goal from the 1-yard line with 26 seconds left, clinching a 28-24 victory for the fifth of Tom Brady’s seven championships.

The game is sometimes regarded as the best Super Bowl of all-time. But watching from the opposite end zone (I was lucky enough to attend on the company dime back when FiveThirtyEight was held in high esteem by my bosses at ESPN1). I remember actually feeling a bit deflated by how quickly it ended.2 According to Sports-Reference.com, plays from the opponents’ 1 result in an interception less than 0.5 percent of the time3 (fumbles are the more common problem because teams usually run rather than pass4). Instead, game over. There were about 2.5 seconds between the snap and when Butler fell to the ground with the ball secure in his hands; even a walk-off home run provides more time for contemplation.

But that game 11 years ago had to live up to a lot of hype. It was a matchup of giants. The Tom Brady Patriots were the Tom Brady Patriots. And the Seahawks had decimated the Broncos to win Super Bowl XLVIII in New Jersey and entered the season as favorites to repeat.

On Sunday, Super Bowl LX — that’s #60 if you don’t speak Roman Numeral — will feature the same uniforms, but there are no players on either roster left over from the 2015 game. And instead of a battle of juggernauts, it’s one of overachievers. At least according to Vegas, whichever team prevails will be one of the least-likely champions ever, right up there with the Rams’ stunning turnaround into the Greatest Show on Turf in 2000 and Brady’s first title in 2002.5 The Seahawks began the season with +6000 odds (60:1) to win the Super Bowl. The Patriots were +8000.

I’d love to tell you that ELWAY gave the Seahawks and Patriots better odds. But I can’t say that because we didn’t get the model up and running until Week 7. I can tell you that it’s been bullish on each team ever since we began publishing our forecasts. Not that ELWAY got everything right (let’s not talk about the Lions, please). But those various ELWAY-inspired Seahawks futures bets I made earlier in the season might be enough to save my year.

Still, the truth is that, as much as it might be “fun” to gamble ELWAY’s reputation on a Super Bowl pick that defied the conventional wisdom, this is one of those games where the model is fairly well-aligned with Vegas. The system has been higher on both participants than the consensus. But since it likes both of them, that sort of cancels out. Still, there’s a lot to consider, including a couple of “X-factors” that aren’t accounted for explicitly by the model.

The history of the Super Bowl, according to ELWAY

Since ELWAY ratings never entirely reset but carry over partly from season to season, that means technically speaking, the very first game in our database from 1920 affects the system’s ratings for every subsequent game. So, we can see how ELWAY would have rated every Super Bowl matchup — and how the Seahawks and Patriots line up against past participants. What I’ll show you here is a simplified version of our retrospective ratings. They include a team’s baseline rating plus an adjustment for the starting QB in the game. There are various other adjustments that ELWAY makes: the most important of these is for injuries. That’s an important one this year, since both the Seahawks and the Patriots are notably healthy (unless Drake Maye’s shoulder problem proves to be something that can’t just be shrugged off). However, we only have injury data dating back for the past few years so I’m not including those adjustments in this table. The other adjustments that ELWAY makes don’t tend to matter much for the Super Bowl.6

Read more

It’s safe to say that this has been a weird NFL season.

Before the year began, Vegas odds posited what were essentially six Super Bowl co-favorites: in alphabetical order, the Bills, Chiefs, Eagles, Lions, Packers and Ravens. Let’s throw in the Commanders here too, since they had the next-best odds and played for the NFC Championship last season.

Of those seven teams, only three even made the playoffs. And only one, the Eagles, is hosting a playoff game this weekend.

True, the standings might have been upended if a few balls had bounced differently. The Chiefs were 1-9 (!) in games decided by seven points or fewer after having been 10-0 (!!) in those games last season. Conversely, the Broncos were 1-6 in such games last season — but 9-2 this year. And that’s before getting into untimely injuries.

ELWAY, the new NFL model that we launched early in the season, didn’t necessarily have a lot of hot takes that bucked the consensus — but it wasn’t a great year for the consensus. Nevertheless, we think ELWAY can offer some real insight on the postseason. By being plugged into QBERT, it recognizes how much of a difference Drake Maye has made for the Patriots, for instance. It can account for the impact of injuries — a big deal in the case of the 49ers, Chargers and Packers — and has more detailed calculations than other systems about the effects of weather, home-field advantage and rest.

And because ELWAY ratings change iteratively over the course of the season, it can weigh preseason expectations — which still matter in a sport where the season is just 17 games long — against who’s coming into the playoffs hot. So here’s how each playoff team’s ELWAY rating has evolved over the course of the year:

(For a more detailed version with non-playoff teams included, see the ELWAY landing page.)

ELWAY started out the year skeptical of the Jaguars, but it has since considerably revised its view of Trevor Lawrence and the rest of the team. The Seahawks and Patriots basically kept looking better and better over the course of the season — and that had been true for the Rams too, until they scuffled a bit in the final few weeks.

Other teams, like the Eagles and Bills, are basically what ELWAY thought they were, though it’s more bearish on both teams than other systems. Speaking of the Bears, ELWAY still stubbornly insists they’re a slightly below-average team, which reflects a combination of lingering preseason skepticism and a defense that gave up 6.0 yards per play, tied for third-worst in the league.

But to answer the question in the headline, who is the Super Bowl frontrunner? (There’s no “favorite” per se since every team’s chances are well below 50 percent.) Well, the photo gives it away: it’s Seattle. But I have a lot more detail for you on how all of this works, with a capsule summary considering the strengths and weaknesses for every remaining team.

The version of the team ratings you see in that pretty chart above doesn’t account for quarterback and injury adjustments, whereas ELWAY’s simulations do. So let’s take at those. Non-QB injuries are often neglected by casual fans, but they go a long way in differentiating positive and negative surprises.

The way ELWAY works, basically every team gets points marked off for injuries. (Our model also accounts for mid-season trades, which can have either a positive or negative impact.) But it’s the relative impact that matters. In the table, the injury/trade adjustment is tantamount to a point spread: the Seahawks project to be only 0.5 points per game worse than if they’d had their entire preseason roster healthy. Other positive surprises like year like the Broncos and Jaguars are also notably healthy.

For the Packers and Chargers, however, injuries take about 4 points off their rating per game. Usually, a team like that is going to struggle to make the playoffs at all: look what happened to the Lions. The 49ers are also coming into the playoffs in poor shape.

One more chart before we roll through the teams. I’ll admit this one is a little busy, but it gets at something critical in a 17-game regular season: the role of skill versus luck.

The Bears had a +22 turnover differential this season, which is highly impressive given that teams are much stingier about turning the ball over than they once were. (The New York Jets literally didn’t have a defensive interception this year.) There is, however, a large body of research suggesting that turnovers are mostly a matter of luck, especially on the defensive side of the ball. Efficiency data — things like yards scored and allowed per play, completions and first downs — tends to have more predictive power. Sacks are a more persistent indicator than turnovers, too.

Four of the 14 playoff teams actually allowed more yards per play than they realized themselves, whereas the Seahawks’ differential was the best in the league. ELWAY’s formulas for calculating all this stuff are somewhat more complex than what I’ve shown above — but the chart gets the basic idea across. An advantage built from a positive turnover differential may be a house of cards.

OK, then, let’s run through the 14 teams who are lucky enough to still be playing football, listed in ascending order of their ELWAY Super Bowl odds.

14. Carolina Panthers (0.2% chance)

Read more

Through an action-packed month, the Houston Rockets are blistering hot. At 10–3, with the NBA’s best offense, they’ve validated their loudest optimists and, so far, silenced their biggest skeptics, with their biggest test since their opening-night OT loss to OKC set to come tonight against Denver.

The Rockets were the NBA’s “surprise team” last season, blowing by Vegas’s preseason win total of 43.5 and climbing to second in a crowded Western Conference before getting bounced in a bruising seven-game series against Golden State. By my estimation, they’ve played their rebuilding hand quite optimally: tanking when necessary, stacking prospects, signing veterans, and capitalizing on their players’ development at the right moments.

Zooming outward, this is really as much as you can ask for since James Harden decided to force his way to Brooklyn five years ago. While some teams that lose a major star spend years bouncing around in NBA purgatory, the Rockets have parlayed a series of smaller moves into a relatively quick path to contention. Houston even made a mediumish-risk, highish-reward splash this summer, trading Dillon Brooks and Jalen Green for Kevin Durant. Durant hasn’t been spectacular, but he’s been healthy and good, and along with Alperen Şengün and Amen Thompson, gives the Rockets three players who rank in the top 35 in the league in EPM.

The Rockets have also embraced some unconventional tactics. Last season, they leaned heavily on Steven Adams’ offensive rebounding to generate second-chance points — the team grabbed an astonishing 43.9 percent of offensive rebounds when Adams was on the floor. This year, they’ve doubled down on size, playing Sengun and Adams in more minutes together. They’ve even supplemented the double bigs with Durant and Jabari Smith, creating further supersized lineups. The result is a 40.6 percent offensive rebound rate so far, a multiple-standard-deviation outlier relative to the rest of the league, and on track to be the highest rate in NBA history.

It’s hard to believe that just five years ago, the Rockets were running full-on Micro Ball with Russell Westbrook and James Harden. But reinvention has basically been their trademark for two decades.

Subscribe now

Some of Houston’s experimentation has been born out of necessity. They don’t exactly have a surplus of shooters, and losing Fred VanVleet to an ACL tear this summer only made that more apparent. On paper, this should be the kind of team built to withstand a rash of injuries — after all, their depth last season was so strong that even No. 3 pick Reed Sheppard couldn’t find consistent minutes. But that depth was concentrated in the frontcourt — the guard rotation is trickier.

Reed Sheppard, sneakily the weirdest player in the NBA

Sophomore guard Reed Sheppard started the year a little flat, but he’s been on a tear over his last ten games — putting up 21 points per 75 possessions on a scorching 69.2 percent true shooting and an even 50 percent from three. Still, the questions about his size and his ability to generate his own offense haven’t gone away. Sheppard is small, but he was not drafted to be a traditional lead guard — he’s much more in the mold of an off-ball player who can add the elite shooting that the Rockets badly needed.

This isn’t coming out of nowhere: Sheppard shot 52.1 percent from long range in his lone season at Kentucky, the sort of eye-popping number that the editor of this story (Nate) initially assumed must be a typo. Although Sheppard mostly came off the bench, it wasn’t a small sample with 144 attempts from three and remarkable consistency across catch-and-shoot attempts, off-the-dribble 3s, guarded and unguarded shots, you name it.

Does the shooting efficiency come at a defensive price? Not necessarily in this case. In fact, much of the intrigue around Sheppard’s draft profile was because of his stocks (steals and blocks) rate of 7.1 percent, abnormally high for someone of his size and (lack of) speed.

The broader question is whether steal or block rates are good measures of defensive impact. There was a point not so long ago when steals seemed like the most important stat. An old FiveThirtyEight article by Benjamin Morris, for instance, found that steals were worth the equivalent of 9.1 points in predicting player impact.

How can steals have such a large impact? Some of it is direct: they end the opposing team’s offensive possession and start a fastbreak. Fastbreak possessions are considerably more efficient. Furthermore, many steals occur on passes that would lead to open shots, or in close-to-the-basket situations where the offense’s expected value is high. So, a steal can be worth several points on its own.

But also, steals are a proxy for overall, harder-to-measure, defensive peskiness and activity. Most of the time, players with high steal rates are also good defenders as measured by on-court/off-court impact stats. So, yes, there’s a lot of empirical evidence that steals are super valuable — but practically, there are some issues with using steals as a measure of perimeter defense, the same way blocks are often used to signal the best rim protectors.

For example, in 2016, Stephen Curry led the league in steals — but how high did he actually rank in defensive impact within the Warriors’ rotation?

Steph certainly wasn’t as impactful as Draymond Green or Andrew Bogut — the Warriors’ defensive anchors — and probably not as valuable on that end as Klay Thompson or Andre Iguodala, who regularly took the tougher backcourt assignments. Curry was typically matched up with off-ball shooters, which freed him to roam passing lanes and rack up steals, even if he wasn’t necessarily the player creating the most defensive “value”.

According to EPM, Sheppard is at −0.3 points per 100 possessions defended so far on the season. EPM is measured relative to league-average — so Reed’s been average, basically. That might feel a little disappointing when you consider his prodigious steal rate (3.3 steals per 100 possessions). But average is more than acceptable for a short, relatively unathletic 21-year-old guard, a profile that typically implies a (massive) defensive liability.

Sheppard is filling big shoes. Why did analytics geeks, including me, think VanVleet would be so hard to replace? It’s a bit of a weird question. FVV last made an All-Star team in 2023.

The easiest answer is that he did some very important “small” things for this very unconventional team. VanVleet has been a positive defender for his entire career. His ability to navigate screens and play up to the ball handler without losing any space has made him one of the best point-of-attack defenders1 in the NBA for a decent stretch of his career, and he’s mostly aged gracefully in that regard.

Sheppard, on the other hand, is playing primarily off the ball defensively because he’s neither very strong nor laterally quick. Still, although he’s limited as a point-of-attack defender and exploitable due to his size, he does have good hands and can play within the gaps while passes are attempted to shooters.

I wonder, though: can Sheppard ever advance from average to being a good or even great defender? It probably depends on his system. It’s telling that the Rockets have already logged 221 possessions of zone coverage through 13 games — the most in the league by a wide margin. (This is becoming a theme: if you look up a given statistical category, the Rockets are probably near the very top or the very bottom.) The zone lets Houston lean more heavily on its double-big lineup. Without VanVleet at the point of attack, they’ve shifted toward a more balanced defensive scheme instead of letting opponents hunt individual matchups. Zones have traditionally been rare in the NBA because they concede more open threes, but they also offer a team-based defensive approach that helps Sheppard play directly to his strengths as a turnover creator.

The curious case of Amen Thompson

The Rockets also boast an electrifying talent in Thompson, who, by my estimation, is the most athletic player in the NBA.

This is not my most controversial take: Thompson is probably the only player in NBA history who can putback dunk his own miss from a standstill and sprint at a top speed of 23 miles per hour. P3, the sports performance firm, categorized him as a “Specimen” — a label reserved for the NBA’s elite athletes. Still, even within that group, Thompson has posted outlier numbers. His athleticism isn’t just raw or flashy; it’s functional. He uses it to sky for rebounds and cover ground without the ball in his hands. That last part —his ability to consume space — is arguably the defining trait of Amen’s game.

We talked a bit in my last piece about “gravity”. So let’s return to the question of why spacing has become so important in the NBA. Many casual fans claim that it’s just because 3-pointers are “worth” more than two-pointers, but let’s think a little bit about that. Is a team’s main objective on every possession to shoot a three? What is the most efficient shot in the game?

Unsurprisingly, rim attempts are still the most valuable shots in basketball. The more important trend, however, is a spike in rim efficiency over the last 10 years.

Raise your hand if you’ve been in this situation. You’re peacefully trying to watch a basketball game, and that one guy — maybe a couple beers down at this point — goes on a tirade about how the NBA is getting softer:

“All players do is shoot threes now! None of them want to get to the rim! Back in the 90s, players played defense!” — Random Guy In Bar2

A simple data dive dispels most of these narratives. We see that rim attempts have been fairly stable since 2001. It’s really just long midrange jumpers that have declined.

It turns out rim efficiency and 3-point attempts are highly correlated, and for good reason: the function of spacing isn’t just that 3 > 2. Rather, spacing makes offensive interactions, such as the pick-and-roll, less congested. In theory, the more functional 3-point shooting a team has, the more space there is to operate, and the offense as a whole becomes more efficient.

These trends also shed light on why many fans think defense is a thing of the past. Offense seems to be “winning” because reliable 3-point shooting now stretches defenses to their breaking point. In terms of pure talent, defenses are probably better than they’ve ever been — it’s just that modern rules and offensive spacing have made it harder for even elite defenders to bend the game in their favor.

I admit this was a long detour, but it leaves us with an essential question for a player of Thompson’s archetype. What happens when a non-big has no hope of becoming a good shooter?

In the mid-2010s, Tom Haberstroh, searching for a way to quantify this, created “respect rating”. Respect rating combined two measures, gravity and distraction, to capture the degree of defensive attention a player commands. In essence, it quantified floor spacing. High scores usually belonged to shooters who stretched out defenses, while most low scores went to those who didn’t draw extra coverage.

The results, for the most part, were not surprising. Stephen Curry, Kyle Korver, Klay Thompson, and other shooters were consistently at the top. However, one name popped up that didn’t belong with the rest: Dwyane Wade.

Why did defenders pay attention to Wade so much, even if he was a pretty famous example of a poor 3-point shooter? You would think, based on some of the narratives about the modern NBA, that non-shooters can’t function outside of a few offensive actions. In D-Wade’s case, it was his cutting. He had a knack for executing his bursts to the basket at the right moments — sometimes unseen and sometimes just too fast — and that required his defender to stay attached to him.

Basketball is a game of attention, and while shooting is a great way to garner it, it’s not a universal requirement. For example, in this play, the Clippers run a pick-and-roll action for Ivica Zubac. Christian Braun camps himself in the paint while Kris Dunn sits at the 3-point line (Dunn is a 32.7 percent lifetime 3-point shooter, despite a diet of mostly wide-open shots). This means the Nuggets have a clear advantage, right?

Well, it turns out the Clippers can scheme around this — Dunn times the cut perfectly, and even if he doesn’t receive the ball, the Clippers are left with an advantage. But Dunn is an okay cutter at best. What happens when the cutter is bigger, faster, and leaps over everyone? When that cutter is Amen — 44-inch vertical, 7-foot wingspan, elite contact finisher — the geometry becomes even more warped. Thompson isn’t D-Wade (yet), but the impact metrics have him as a significant net positive on offense (not to mention a DPOY contender) largely because of that cutting.

Şengün is pressing all the right buttons

Part of Thompson’s upside depends on the personnel around him, particularly Şengün. I’ll be up front: I came into this season as a Şengün skeptic. Some of this was perhaps grounded in aesthetic bias: I’m not opposed to post players so much as doubtful of their ultimate ceiling.

Furthermore, the “Baby Jokić” comparisons weren’t exactly positive to me. Nikola Jokić is a player who fundamentally shouldn’t work — but he’s synergized skills that stretch the imagination. When most fans think about Jokić, they’re probably reminded of the ridiculous post-up scoring and passing. But is that the only way to describe his uniqueness? Geometrically, Jokić also provides value from his ability to occupy areas other than the low block. It’s not just that he can score from the post, it’s that he allows possessions to flow downstream without disrupting the tempo of his teammates’ cuts.

This ability has only manifested in a few players in NBA history — Jokić, Chris Webber, and Vlade Divac are among them. Despite being a good post player, Şengün had struggled with the rest of that skill set. Last season, he didn’t grade out as efficient across most play types, especially those that depended on athleticism. As a roll man and cutter, he was one of the worst centers in the NBA:

Granted, the sum was somewhat greater than the parts: the Rockets had a +7.5 net rating with Şengün on the floor last season. But this year, that’s exploded to a staggering +13.4 with him on the court.

Is this early-season noise or the result of real development? One notable finding is that Şengün’s share of possessions as the roll man in PNRs has declined significantly — from 18.2 percent in the 2024-25 season to only 6.4 percent now. While some of this is the result of their offseason moves3, it’s possible the Rockets are pivoting entirely towards a Şengün-ball offense.

And defenses have reacted with more attention. Last season, the defense committed to or double-teamed Şengün on 65.1 percent of his post-ups4 — a solid rate, but nothing in comparison to players like Jokić or Giannis Antetokounmpo, who saw rates above 75 percent. This season, Sengun’s double-or-commit rates have skyrocketed to 80 percent in an admittedly small sample size. The efficiency of his postups and isolations has increased as well, suggesting real development.

What’s the final verdict?

In some ways, the story in Houston isn’t that complicated: young players typically improve, whether incrementally or through bigger leaps. As we’ve covered, Şengün, Thompson and Sheppard are unusual players — but they’re increasingly making the most out of their roles. The odds for that sort of positive development are much more favorable for younger players, but it also speaks to the good work of the front office and coach Ime Udoka.

There are also more red flags here than for a typical 10-3 team. It’s early in the season, and the league will find counters to the Rockets’ unusual tactics — if not now, then in the playoffs. Nobody on the roster is a safe bet to make one of the three All-NBA teams. VanVleet’s injury is obviously not ideal, and Houston might be best served by taking a more egalitarian approach to ball-handling this year. Letting the offense take shape organically while players continue to grow into their roles feels like the right move for a roster still figuring itself out.

Even so, this is one of the league’s best teams. I don’t think they’re equipped to beat Oklahoma City just yet — to be fair, who is? But they’re moving toward the inner circle of contenders, and there is the shape of even better things to come — a leap that should have H-Town buzzing.

This article is free. But if you’re an NFL nerd, I’d strongly encourage you to sign up for a paid subscription for full access to ELWAY and QBERT. They’re a lot of fun, especially as we’ve continued to add new charts and data. And fingers crossed, but so far ELWAY has also been ahead of schedule at keying in on the success of teams like the Seahawks and Patriots.

Subscribe now

I’ve never exactly not been an NFL fan. When I was 15, I even ran the equivalent of a neighborhood football betting pool.1 It would have been nice if those Lions of my youth had given Barry Sanders more than one playoff win.

Because of the league’s ubiquity in American culture, though, the NFL is the background hum that pervades every sports fan’s life. You’re always hearing it, seeing it — at the bar, the airport, the poker room — without necessarily really listening to or watching it. But suddenly, because of ELWAY and QBERT, I’m an avid fan of the NFL again. As with visiting a city where you lived as a kid, there’s an uncanny familiarity — that pizza place is still open? — but you also notice all the little changes.

For instance, with a fresh set of eyes, it’s easier to see why the NFL remains such a popular product with the modern, episodic viewer2 even as other sports have struggled to hold their audience. But it’s also a much different league than my teenage archetype. Take, for instance, the proliferation of final scores that read like high school locker combinations. For example: 36-29. 44-26, 44-22 (twice!). And that was just last week alone. These are not the scores of our pastoral football childhoods: the familiar 28-24s, 20-17s, and 21-13s generated piously by some sensible linear combination of 7-point touchdowns and 3-point field goals.

Of those scores last week, only 36-29 was officially a Scorigami, the term invented by the sportswriter and humorist Jon Bois for the first appearance of a score in NFL history. That’s because Scorigamis are a nonrenewable resource. What were once vast, untapped deposits of Scorigamis have gradually been strip-mined away, particularly since the introduction of the 2-point conversion in 1994. The remaining uncharted territory on the Scorigami map now requires dedication and creativity to reach, such as the 40-40 tie between the Cowboys and Packers or the Bills’ spectacular 41-40 comeback win over the Ravens in Week 1.

Nevertheless, I’m here to expand and revise the definition of Scorigami. 36-29 is still a scorigami even if it occurs again next week, I decree. You’ll notice I used a lower-case “s” there. A Capital-S Scorigami is the first instance of a particular score in NFL history. But a scorigami is any sufficiently weird, profane score, whether 2-0, 39-26 or 40-40.

And there are more and more of these. In the early days of the league, 77 percent of final scores were one of the following: 0, 3, 6, 7, 10, 13, 14, 17, 20, 21, 24, 27, 28, 31 or 34 points. Those represent every combination of zero to four touchdowns (with a successful 1-point conversion) and zero to two field goals. Now, only slightly more than half of final scores are one of those totals. The number 68 might have been the yardage on Jaguar kicker Cam Little’s recent record-setting field goal. But if an NFL game ends with a final score of 68-11 this Sunday, it honestly wouldn’t feel that weird.

Thus, I’m here to dub 2025 as the dawn of the Scorigami Era in the National Football League.

Share

There is, for one thing, a lot of scoring. The new NFL is prolific. At 46.6 points per game3, this year is tracking to be the third-highest-scoring season in NFL history (and, for what it’s worth, offenses generally improve as the season wears on). Everybody is getting in on the fun. While passing yardage has declined from its peak, quarterbacks are contributing more from the rushing game than ever. Kickers are making the impossible look routine. Rushers are averaging 4.4 yards per carry this season, close to the all-time high of 4.5 YPC from 2022. And kickoff returns are a thing again.

But as with origami, the second half of Bois’s portmanteau, the new NFL is precise. Coaches are finally nailing intricate 4th-down and 2-point-conversion decisions, long the bane of every NFL stat nerd’s existence, and strategy throughout the drive is changing in anticipation of a conversion attempt. Meanwhile, teams are stingier than ever about turnovers, and drives are longer as offenses efficiently string together a series of low-risk, medium-reward plays.

And yet, the net effect of the Scorigamified NFL is a little peculiar, as we’ll see.

NFL offenses are even more prolific than they seem

In working on projects like QBERT, which tracks all QB performances since 1950, and ELWAY, which dates back to 1920, you develop a sense for when there are phase changes in the sport. Some of these are even hard-coded into the models in various ways. When you’re working on a model, there’s often a trade-off between drawing the window further back because it gives you a larger sample size, or essentially chucking out the “old” data because it describes a regime that no longer exists in quite the same form.

One cute little thing we do in ELWAY, for example, is simulate discrete final scores. The raw form of the algorithm, for instance, might draw from a probability distribution to simulate a “score” like Chiefs 29.1, Broncos 22.6. Obviously, you could just round those numbers off: Chiefs 29, Broncos 23 is a perfectly plausible score. However, even in the Scorigami Era, scores like 30-23 or 28-24 or 28-21 are considerably more likely. (Not to put too fine a point on it, but these small differences matter when projecting the likelihood of a given point spread being hit, for instance.)

I won’t bore you with all the details, but essentially, ELWAY solves this problem by drawing from a database of 1 million simulated games played under our best estimates of modern NFL conditions. Each team gets a discrete number of drives, which might result in 0, 3, 6, 7 or 8 points — or occasionally, 2, 6, 7, or 8 points for the other team in the event of a safety or a defensive TD.

This is trickier than you might think because the home and away teams’ scores are interrelated. Especially late in the game, the current game score profoundly affects strategy: teams with leads run out the clock while trailing teams go for broke. The object of the game isn’t strictly to maximize the number of points scored, but to score at least one more than the opponent. In our simulations, a team trailing 21-13 late in the game will never kick a field goal on its final possession, for instance, and it will always go for two if it’s lucky enough to score a touchdown.

Still, the backbone of ELWAY’s simulations is basically this:

Excluding possessions where the clock runs out on the half, these are essentially the eight ways that an NFL drive can end. So it’s important to know the relative frequency of these outcomes. And that’s been changing a lot. As compared with the start of the millennium from 2001-2005, teams are scoring considerably more touchdowns and field goals (but missing slightly fewer FGs as kicking accuracy improves). They’re also punting less and committing fewer fumbles and interceptions — although they’re turning the ball over on downs more often because of a much higher propensity to go for it on fourth.

Overall, the average drive so far in 2025 is producing slightly more than 2.1 points, as compared to 1.6 points over the 2001-2005 window. That’s roughly a 33 percent increase. And that’s huge. Larger than the 28 percent increase between when the NBA bottomed out at 91.6 points per game in its 1998-99 season to its high-wattage 117.0 PPG (!) so far this year.

But there’s a “trick” that keeps this offensive explosion hidden. In football, successful drives run more time off the clock: a 17-play, 88-yard scoring drive might take up fully half the quarter, while a four-and-out can occupy less than a minute of game time, especially since incomplete passes stop the clock while completions usually do not4. So far this year, each team is averaging 10.5 drives per game, compared to 12 drives per game as recently as 2013. Thus, even though overall scoring has increased, the net increase in final scores considerably understates the gains in offensive efficiency.

Where do these gains in scoring efficiency come from? One important factor is that the field has essentially been shortened on both ends.

So far this year, following opposing kickoffs, teams have taken possession of the ball a tick downfield from their own 30-yard line on average. And this is the best starting field position they’ve had in years. The reason is the NFL’s dynamic kickoff, introduced last year. The new kickoff was designed to incentivize returns while reducing the possibility of injuries. In 2023, the last year of the traditional kickoff, 73 percent of kickoffs went for touchbacks because most modern kickers can boot the ball deep into the end zone, and returning teams generally wouldn’t bother with the marginal cases anyway lest they risk an injury.

Last year, 64 percent of kickoffs resulted in touchbacks all the same, even though the touchback was moved to the 30-yard line. But this year, the touchback has been moved out further to the 35-yard line, reducing the touchback rate to just 17 percent. Interestingly, this further tweak has only improved field position by about half a yard on average; kickers are becoming more adept at finding the corners of the landing zone. Still, possessions following kickoffs are starting around 7 yards further downfield than they did a decade ago.

If you’re starting at, say, the 33-yard line, you’re now really only two first downs away from being within field goal range because field goal range basically starts at midfield. As I’ve learned the hard way — I’ve, uh, lost a few bets recently this way — kicking off at the end of the half with say 20 seconds left on the clock is suddenly quite dangerous. The offense starts with the ball somewhere around its 30 … two quick completions … 59-yard field goal attempt … boom, game (probably) over.

Kicking efficiency has been improving for years, but relaxed restrictions around so-called K-balls this year — kickers basically get to use a fresh-out-of-the-box ball instead of one scuffed up by game action — has been an accelerant. If kickers like Little (who also kicked a 70-yard FG in a preseason game this year) and the Cowboys’ Brandon Aubrey are exceptions, it’s only by a few yards’ worth of distance. The rate of field goal attempts of 55 yards or more has increased by 135 percent compared to just four seasons ago, in 2021. And the number of makes from 55-plus has tripled in just four seasons. I’m guessing we’re in for some mean reversion — but so far on the year, kickers are actually converting 64 percent of their attempts from super-long-range.

But let’s not give too much credit to those kickers. I’m of the view that the value of quarterbacks, if anything, is a little bit understated by the conventional wisdom.

Quarterbacks aren’t quite putting up the gaudy counting statistics that they did a few years ago. Despite the 17-game schedule, no QB has passed for 5000 yards in a season since 2022, and none are on pace to do so this year.

Part of that is that QBs are opting for shorter, safer routes with exceptionally high completion percentages and little risk of interceptions. The average number of yards per completed pass is 10.1 this year, which is tied with 2023 for the lowest in NFL history.

However, advanced statistics like QBERT suggest that this risk aversion is probably smart for once. The league’s collective unadjusted QBERT rating is 89.6 this year, which is the second-highest ever behind an outlierish, pandemic-affected season in 2020.5 And in traditional NFL passer rating, quarterbacks are tied for their highest-ever rating at 93.6.

QBERT and passer rating differ ever-so-slightly in how they approach moving the ball downfield versus efficiency — QBERT is actually slightly more forgiving to old-school, gunslinger-type QBs — but it’s not a big difference. However, QBERT also accounts for a QB’s rushing contributions, as well as fumble and sack avoidance, whereas traditional passer rating does not.

Having, at the very least, some scrambling ability is now the norm rather than the exception. Thus far in the 2020s, starting quarterbacks have averaged 316 rushing yards6 per 17 starts, nearly twice the yardage they accumulated in the 1980s and 1990s. For the first time in NFL history, quarterbacks as a collective are gaining enough rushing yards to outweigh the yards they lose from sacks (even though this has also been a good era for pass rushers).

QB rushing touchdowns and first downs have increased proportionately. The latter category is particularly important; QBERT really likes quarterbacks who run for first downs. (Which helps to explain its Jalen Hurts fetish, for example — in 2023, he was second among all players in rushing first downs, including running backs.) The reason is that these plays produce a big gain in expected value. They often come on third down, and successful scrambles often reflect bailouts from what would otherwise have been negative-yardage sacks. The recent going rate of about 1.4 QB rushes for first downs per start might not sound like a lot, but those are often drive-saving plays.

Strategy is becoming more precise, especially on 4th down

In 2013, the New York Times launched its 4th Down Bot. It dutifully analyzed every 4th-down play and found, in line with a long lineage of work by academics and independent analysts, that teams were much too conservative, punting and kicking field goals when the rewards outweighed the risks if they’d simply entrust their offense to pick up a few more yards. 4th Down Bot last tweeted in 2017, claiming it was “tired” and moving on from the project.

Honestly, I’m relieved that our robot overlords still need days off. But I can’t exactly blame 4th Down Bot for feeling frustrated. Up through 2017, there had been no increase at all in 4th-down attempts despite all this research; if anything, they’d been declining slightly, perhaps due to some high-profile “go-for-it” calls that didn’t work out so well.

But like an artist who only becomes famous posthumously, 4th Down Bot, if it were resurrected today from its final resting place in a GitHub repository, would find that it hadn’t been wrong, only ahead of its time.

In 2017, NFL games averaged 1.8 4th-down attempts and 0.8 successful conversions. However, so far this year, those numbers have ballooned, roughly doubling to 3.1 and 1.8, respectively.

Note that this data is for both teams combined, so each team has approximately one successful 4th-down conversion per game in today’s NFL. Still, that goes a long way toward explaining why teams are stringing together longer drives. The success rate on 4th downs is also increasing. After consistently hovering around 50 percent for most of the league’s history, the 4th-down conversion rate has improved to 59 percent over the past two seasons.

The increased success rate is partly due to the tush push and other QB sneak plays. On fourth and very short — one yard to go or less — teams are opting to run about 75 percent of the time, and they’re converting 75 percent of those rushes as compared to 55 percent of pass attempts.

The data strongly suggests that teams should be running more; rushing plays are also converting at a higher rate than passes on 4th-and-semi-short (2 or 3 yards to go) and even at longer distances through the occasional draw play (although the sample sizes are small). At the same time, the precision of the short-yardage passing game is still enough to prevent the defense from keying in too much on the rush.

Other than mixing in even more runs, there may not be all that many gains left to be had from improved 4th-down strategy. In Week 10 games as analyzed by ESPN Analytics — basically the modern successor to 4th Down Bot — teams made the algorithmically correct decision 84 percent of the time on 4th down last week on plays where there was a material amount of difference in win probability7. And a handful of the “incorrect” decisions — mostly by the Philadelphia Eagles — were actually disliked by the bot for being too aggressive rather than too conservative.

Planning ahead for 4th-down attempts can also make the rest of the series more efficient. Compared to historical baselines, teams are now passing more on 2nd down at all yardages, but are actually running (slightly) more on 3rd down.

This makes sense if you work backward through the decision-making process. Take a typical 2nd-and-6 or 2nd-and-7 situation. If a team has only two downs to play with because it’s going to kick on 4th, it may figure it has to run or else it’s basically taking the running game out of the equation for the rest of the series (6-7 yards is a lot to ask for from a rushing play).

From a game theory standpoint, it’s always costly if your opponent knows what type of play you’re going to run. But if you’re planning on going for it on 4th, running on 3rd-and-semi-long situations (3rd-and-6, 3rd-and-7) can potentially set up a 4th-and-short. Meanwhile, on 2nd-and-short, you now have basically a free play downfield since you’re almost certain to convert given two opportunities to do so.

That’s especially true for teams like the Eagles that succeed in short-yardage situations at exceptionally high rates. They’re now basically facing a 1st-and-9 rather than a 1st-and-10 at the start of every series; that’s a huge advantage, although maybe one that ELWAY and QBERT are having a little trouble grasping.8

The rate of 2-point conversions has also increased, although progress in this direction has come in fits and starts. When the 2-point conversion was first introduced in 1994, teams played with their new toy fairly often. But the number of 2-point attempts per game fell from 0.25 per game in the debut year to just 0.07 by 2006.

Moving the 1-point conversion (extra point) line back to the 15-yard line in 2015 inspired more aggression because an extra point was no longer as much of a sure thing. Since 1994, 2-point conversions have had a 48 percent success rate (and this hasn’t really changed much9), thus producing 0.96 points per play. In terms of expected value — granted, not always the most important metric since the particular scoring margin matters a great deal for optimal strategy — the 2-point conversion produces fewer points if a team converts 99 percent of its extra point attempts. But if it only makes 95 percent of its extra points, about the rate since the rule change, the 2-point conversion results in slightly more points on average.

Why are there more weird scores?

The combination of more 2-point conversions and occasional missed XPs has a profound effect on the incidence of scorigamis. So far this year, 14 percent of touchdowns produce either 6 or 8 points, rather than 7. With 5.3 touchdowns per game combined between both teams, that means the typical game produces at least one TD that doesn’t result in adding 7 points to the scoreboard.

Granted, “typical” is doing a lot of work there, because unconventional conversion situations can come in bunches. If a team misses an XP and thus trails by 8 points prior to scoring, it will be much more likely to go for two following a touchdown. Still, this can contribute even further to funky scores. Say a team trails 17-9 and scores a TD. It will probably go for two. But if it fails to convert, it will be sitting on 15 points and is then basically past the scorigami event horizon for the rest of the game, with final scores like 18, 22 or 29 points becoming likely.

Even without any of this 2-point conversion business, the proliferation of field goals has contributed to scorigamis. So far this year, there have been 36 games where a placekicker made three field goals in a game, along with 18 instances of 4 FG games and six where they made 5 FGs. Overall, teams now score three or more FGs about 20 percent of the time, and the resulting scores are inherently scorigami-friendly. Three field goals produce 9 points, which is the same as a TD (with a plain-vanilla extra point) plus a safety.

A final factor in the scorigami singularity is simply the overall increase in scoring. Nearly all the capital-S Scorigamis that have filled in since 1994 have been high-scoring games. From the dawn of professional football through the start of the Super Bowl Era in 1966, the most common final score was actually 0 points. In fact, it was more common for a team to finish with exactly 2 points — so a safety and nothing else — than relatively common modern scores like 29 points or 40 points. Since 2000, only 10 Scorigamis — the exceptions are 58-0, 28-2, 36-3, 47-3, 52-3, 24-5, 31-5, 37-5, 38-5 and 52-6 — have been produced by the losing team scoring 7 or fewer points.

On the higher end of the range, however, new NFL overtime rules are providing additional Scorigami opportunities as each team is now guaranteed at least one possession under most circumstances.10 A game that ends 35-35 in regulation was once very likely to result in either a 38-35 or 41-3511 final score, neither of which is a Scorigami. However, the new rules open up additional possibilities like 43-42 or a 42-42 tie.12

Let’s compare four distributions of final scores:

• The empirical distribution in what I’ll call Early Pro Football from the founding of the APFA (the predecessor of the NFL) in 1920 through 1965;

• The Classic Super Bowl Era from 1966 through 1993;

• The Modern Super Bowl Era from the introduction of the 2-point conversion in 199413 through today;

• And finally, the simulated distribution of final scores from ELWAY, which is calibrated based on our best estimates of current scoring rates, strategies and overtime rules.

In Early Pro Football, 77 percent of final scores were one of the canonical scores I mentioned earlier — 0, 3, 6, 7, 10, 13, 14, 17, 20, 21, 24, 27, 28, 31 or 34 points, everything that results from 0-4 TDs (with an extra point) and 0-2 FGs. The rate of canonical scores declined only slightly to 72 percent in the Classic Super Bowl Era. But it has been just 62 percent so far in the Modern Super Bowl Era, and we expect that to decline further to 55 percent going forward based on ELWAY’s simulations of current conditions. There is an even more profound effect on both teams finishing with unusual scores.14

So what part of Scorigami Territory is likely to be conquered next?

The most likely unrealized Scorigami is 36-23 – honestly, not that weird a score — which has about a 1-in-1400 chance of occurring in any given game according to the ELWAY sims. Among scores where the losing team scores a touchdown or less, the most likely options are 27-8, 39-6 and 29-8. Among ties, your best bets are 25-25 and 22-22.

Overall, the chance of a Scorigami is only 2.7 percent in any given game, which means the over/under line is roughly 3 Scorigamis for the rest of this season. But in the Scorigami Era, every NFL game has the potential to produce feats of prolific and peculiar scoring that you’d never have imagined in your childhood.