NBA Eastern Conference preview
Projected standings, plus information that might be used for gambling purposes
The start of the NBA season is two weeks away. It ought to be one of the most competitive seasons in a long time. But I feel a little naked this year — if also a little liberated.
For the time in several seasons, I’m without the RAPTOR projections that we usually published about this time of year at FiveThirtyEight. Along with the other sports models that I authored, they’ve gotten caught up in my long-term planning process, which is still underway. I love these sports models, and am in active conversations about a home for them, whether it’s selling or licensing them to a third party, or publishing them at Substack somehow. But that process isn’t going to be completed in the next two weeks, needless to say.
So instead, I’m going to do something a little different: come up with projected standings on my own. In certain ways, this might be more fun anyway. There are always things that are tricky for algorithms, such as considering team effort levels and incentives to tank, the possibility of mid-season trades, and unusual players that I might have a different subjective view of than RAPTOR does.
I’m a big NBA fan and bet the league on a game-by-game basis last season as part of what was somewhere between a side-hustle and an experiment for my book. I learned a lot from betting, though I soon got limited by a number of the major US retail sites. I came out ahead, although only modestly. RAPTOR projections were a part of that process, but for many reasons — such as that we weren’t always up-to-date on injury info — it wouldn’t have been a good idea to follow them blindly. So I’m used to looking at NBA numbers and then making a lot of mental adjustments to them. I’m curious to see how I’ll do without RAPTOR to anchor me, although there are other projections out there such as Kevin Pelton’s numbers. For what it’s worth, I am actually betting many of these projected records versus Vegas projected team win totals to have some skin in the game for this column.1
Teams are listed in their projected order of regular season-finish. There’s a bunch of info listed for each team; let’s use our friends the Washington Generals as an example:
Washington Generals: 9-73 🟢
2022-23: 8-74 (10-72); missed playoffs; 30th offense, 30th defense
2021-22: 12-70 (13-69); missed playoffs; 28th offense, 30th defense
2020-21*: 10-72 (9-73); missed playoffs; 30th offense, 30th defense
Key additions: none
Key subtractions: Red Klotz
Weighted average age: 25.2
Vegas line: 10.5 wins†
The big headline number associated with each team is my projection for their W-L record. There’s also a color icon designating my confidence level. Green (🟢) means high confidence, yellow (🟡) means medium confidence, orange (🟠) means low confidence and the red flag (🚩) means there’s a special circumstance that makes it hard to make a projection at all.
Next, for each of the past three seasons, I’ve listed a team’s actual won-loss record, and in parenthesis, it’s expected W-L record based on its point differential. (In general, expected W-L is a better predictor of future team performance.) For 2020-21, which was shortened by the league’s recovery from the COVID bubble season, these records are prorated to 82 games (the asterisk serves as a reminder of this). I’ve also listed each team’s ranking among the 30 NBA teams in points scored and allowed per possession.
Key additions and subtractions are players who are joining or leaving a team’s rotation since the end of last season. What counts as “key” is a little subjective and I’m not claiming to be comprehensive.
Weighted average age is based on depth charts from the truly excellent fantasy site BasketballMonster.com. Specifically, they’re weighted based on projected minutes played, although with an extra bonus2 to players in the starting lineup.
Finally, the “Vegas” line is based on a consensus of US retail betting sites. If there is a lot of disagreement in the lines, that is indicated with the dagger (†) symbol. Although I’ll talk explicitly in some cases about whether I like the over or under, please make sure to check for the latest line — and to shop for the best line — if you’re thinking about betting these. Edges are thin, and I’d encourage you to be careful especially for teams designated as 🟠 or 🚩.
All right, that’s more than enough preamble. Let’s go! My projected Eastern Conference regular season standings are as follows. I’ll put the first two teams in front of the paywall but (sorry! this was a lot of work!) the rest is for paid subscribers.
Boston Celtics: 56-26 🟡
2022-23: 57-25 (57-25); 11-9 playoffs; 2nd offense, 3rd defense
2021-22: 51-31 (59-23); 14-10 playoffs; 4th offense, 4th defense
2020-21*: 41-41 (44-38); 1-4 playoffs; 10th offense, 14th defense
Key additions: Jrue Holiday, Kristaps Porzingis
Key subtractions: Marcus Smart, Robert Williams, Malcolm Brogdon, Grant Williams
Weighted average age: 28.2
Vegas line: 54.5 wins
It’s been quite the roller-coaster in Boston. The Celtics experienced a dramatic turnaround two seasons ago, starting out the year 16-19 before playing some of the best basketball we’ve seen since Peak Warriors for the second half of the 2021-22 season. Then they went ahead 2-1 in the NBA Finals against the actual Warriors. Then they blew it, losing four straight. Then the Coach of the Year, Ime Udoka, was suspended for a year for having an inappropriate relationship with a female Celtics employee and eventually found his way out of town. Then the C’s had a very good regular season last year despite that. Then they fell down 3-0 against the Miami Heat in the Eastern Conference Finals. Then they nearly became the first NBA team to complete a comeback from that deficit. Then they shat the bed in Game 7. And now they’ve made two very big trades in the off-season.
RAPTOR (in)famously loves the Celtics. I had bets on Boston to win the Eastern Conference (yay!) and NBA Finals in 2021-22 (oops) as well as in 2022-23 (oops). So my inclination, of course, is to bet the over again. And indeed, RAPTOR likes the Celtics’ additions more than their subtractions, with Holiday (+5.7 points per 100 possessions last year) and Porzingis (+4.8) rating ahead of Smart (+1.4), Robert Williams (+1.7), Brogdon (+1.2) or Grant Williams (-1.3). On paper, it’s a hell of a top six. With the line at “just” 54.5 wins, it’s a tempting over bet.
Of course, there are obvious issues. This is a lot of change for a contending team. It might take the Celtics time to gel, especially under Udoka’s replacement, the still-inexperienced Joe Mazzulla. The team is now fairly old and injury-prone, and there’s a steep drop-off after the top six — though the Celtics do have a smart front office and enough draft capital to add more pieces mid-season.
It’s also the case that Vegas win projections have a history of being overconfident. Teams projected to win more than about 55 games have historically been good under bets, and teams below 25 games have been good over bets. Now, this is one of those patterns where once people notice it, the market may adjust and the edge goes away. The broader point, however, is that winning more than about 55 games in an NBA regular season is pretty hard; you can’t afford to have much go wrong. I just think the top-end talent is so good here — that the upside case is 60+ wins — that perhaps the Celtics can even tolerate a slump or an injury at some point and still hit this number.
Milwaukee Bucks: 50-32 🟡
2022-23: 58-24 (50-32); 1-4 playoffs; 12th offense, 4th defense
2021-22: 51-31 (49-33); 7-5 playoffs; 3rd offense, 14th defense
2020-21*: 52-30 (55-30); 16-7 playoffs; 6th offense, 10th defense
Key additions: Damian Lillard, Adrian Griffin (coach)
Key subtractions: Jrue Holliday, Grayson Allen, Jevon Carter, Mike Budenholzer (coach)
Weighted average age: 30.2
Vegas line: 54.0 wins†
Before the Damian Lillard trade, the Bucks were one of my more emphatic under picks. I really like that trade for them. But I think of it as more of a floor-raiser than a ceiling-raiser. So I still like the under here.
For starters, note that the 58-24 record last season probably anchors people too high on the Bucks. The team’s expected W-L record based on their point differential was only 50-32. And of course, there was their disastrous exit to the Miami Heat in the playoffs.
The Lillard trade ought to solve the problem of the Bucks’ stagnant half-court offense. That’s a big thing to fix. But, especially with Allen also having been moved in the deal, it contributes to another problem that is highly pertinent in the regular season: depth. There is a very big drop-off between the Bucks’ top 4 — Lillard, Giannis, Brook Lopez and Khris Middleton — and everyone else. And this is not a franchise who has shown a Miami Heat-like ability to pick up useful pieces on the margins.
It’s also an old top 4. Lopez will be 35 this season, Lillard 33, Middleton 32 (and started only 19 games last season). Giannis is 29 and hasn’t been close to an 82-game player in some time. He’s coming off surgery and will get probably 5-10 load maintenance days off and miss another 5-10 games due to short-term injuries. The Bucks’ cupboard is very bare as far as having draft equity to improve their rotation.
The top-end for this club is high. The Giannis-Lillard duo is possibly the league’s best. By the playoffs, the Bucks may have have stumbled upon a buyout player or an unexpected internal solution that can improve the team’s depth. But an 82-game regular season is a different proposition.