The NBA gambling scandal, explained by an actual gambler

43 thoughts about the biggest betting scandal in years. Not only is it not surprising, but it looked suspicious in real time.

The Charlotte Hornets’ Terry Rozier loses the ball in a December 29, 2023 game. Getty Images.

We’re just four days into the NBA season, and there have already been a lot of fun headlines, like Steph Being Steph, Victor Wembanyama looking like an MVP, and a double-OT thriller in a 2025 Finals rematch on Thursday night. But whenever your sports league makes front-page news in the New York Times, it’s probably bad news.

I love the Times (and used to work there). But it reserves its front-page real estate for stories it considers of profound international importance (so, the World Cup, yes, but the Super Bowl, no). It takes something truly monumental for an American sports league to receive the Page One treatment three days into its season. Like, say, a sports betting scandal involving the mafia, two federal indictments, and the arrest of an NBA player and head coach.

I have lots of takes on this, informed by my experience covering gambling and being a sports bettor myself. In fact, much of the alleged suspicious activity occurred during the 2022-23 NBA regular season, the year I bet the league every day as part of a sort of experiment/side hustle for my book. After a great start to the season, my trajectory was rocky; I finished in the black, but not by much. Betting the NBA regular season turned out to be a grind, mainly for one reason that also figures prominently in the indictments: the constant ambiguity surrounding star players’ availability due to minor injuries, “load management,” or tanking.

When betting the NBA, a difference of a single point — say, Vegas has the Miami Heat favored by 3 points, but you think they should be favored by 4 — is enough to turn a losing bet into a winning one, or vice versa. But the availability or lack thereof of a LeBron, a Steph, or a Jokic can shift the point spread by 6 points, 8 points, or even more. It is extremely valuable to have inside information about who’s actually playing — the sort of info that the alleged conspirators had.

Without that, your choices are between making a negative-EV bet, sitting the game out, or trying to read the tea leaves of where the insider money is flowing. The detailed notes I kept in 2022-23 show that I often did find something suspicious with betting lines for games mentioned in the indictment, though I didn’t always make the right bets. Sometimes I made the classic gambler’s mistake, being tempted by a line that looked too good to be true — and it was, because the insiders knew something that I and the rest of the public didn’t.

The second federal indictment involves poker, another subject I obviously know well. I actually have less to say here because the illegal underground games described in the indictment don’t overlap so much with my part of the poker world. But I can at least walk you through some of the distinctions between your friendly neighborhood home game and what the Feds alleged occurred in Manhattan, Las Vegas and other places. Hint: you probably shouldn’t be eager to play in poker games featuring a random athlete or two, and also people with nicknames like “Black Tony” and “Albanian Bruce”.

Some obligatory disclaimers

News coverage of this story is also fraught because not only are the sports leagues in bed with the sports betting companies: so are many of the media outlets that cover them. ESPN has its own branded sportsbook, for instance. The Athletic (owned by the NYT) and The Ringer have sponsorships with sports betting companies, too.

As for Silver Bulletin, I’ll proudly acknowledge that one of the use cases for models like ELWAY, our new NFL projection system, is for people considering betting on the action.¹ We explicitly recommend against using our models to bet without also having strong knowledge of both the sports themselves and the practical mechanics of sports betting — the margins are slim, and the books take a cut of every bet, so you need to get pretty much everything right to beat the house.

But I obviously have no ethical qualms with people who want to beat the lines the right way, i.e., with smart modeling and general sports knowledge based on publicly available information. And as I’ve said, I bet on sports myself. The bets are a small fraction of my income, partly out of choice (I have better ways to make more money with less risk) but also because even if I wanted to bet more, I’m limited in some capacity by most of the US-facing sportsbooks. There’s nothing special about me: they don’t like people who are even trying to be sharp. The combination of these limits on the one hand, and already-small edges being chipped away at by bettors with inside info on the other hand, makes the dream of living a good life as a sports-betting sharp off limits to all but a small fraction of people.

It’s early in the life of this story, so the rest of this is going to take the form of a series of bullet-pointed thoughts. I’m reserving the right to reprise this into a narrative-type story once more of the facts are known.

Overarching thoughts

1. Number 1 with a bullet is that I’m not that surprised by any of this. And I’m not just saying that to sound cool or world-weary. Because individual players have such a significant impact, the NBA is particularly vulnerable to cheating based on inside knowledge of player availability. Player prop bets — will Terry Rozier have more or fewer than 4 rebounds?— are another known vulnerability because of the possibility of collusion between gamblers and players.

2. I’m also not surprised by the growing backlash against sports betting, as seen in public opinion surveys. Gambling in general, and gambling on sports in particular, has usually existed in a liminal space. It’s been tolerated at certain times and places, whether explicitly or via some degree of “don’t ask, don’t tell”. But sports gambling has rarely been as “in your face” as it has been in the major American sports leagues for the past few years. There’s a lot of prudishness in the media about sports betting, which isn’t to say there aren’t also real harms. Still, ultimately, states are responsible for regulation, which means that public opinion matters. The continued expansion of sports betting into more states isn’t necessarily something I’d bet on — and it wouldn’t surprise me one bit if the dial moves the other way.²

3. You can make a pretty good defense of the licensed sportsbooks in that they’re better equipped to catch cheating than their shadier offshore competitors or traditional bookies. They have KYC practices — they know, more or less, who’s betting on what — and they can aggregate a lot of data together to look for suspicious betting patterns. And they have both incentives and requirements to work with regulators and law enforcement. It’s likely that some of this stuff has been going on for years, and now it’s easier to detect.

4. Still, I don’t find myself with much sympathy for the big US sportsbooks. While I respect their right to exist and have friends who work for these companies, I think they’ve been short-sighted in various ways, such as by offering too broad a menu of bets (some of which are more subject to cheating and manipulation), by so aggressively limiting skilled bettors, and through advertising that is so widespread as to understandably produce a backlash, especially on college campuses.

5. While there has always been a penumbra of gambling in the NBA — poker games in the back of the team plane, big wagers on golf, legal gambling in some states and illegal bookies in others — I also assume that the explicit endorsement of legalized gambling in the NBA and other leagues creates more of a permission structure for players and team personnel who might otherwise think twice before doing something foolish.

6. There are no free lunches here. Sensible constraints and regulations are very likely to reduce the profitability of the licensed sportsbooks.

7. However, there’s also another obvious trade-off. Restrictions on options in the regulated American market will push some of the action toward offshore sites, which are harder to monitor than ever because deposits are often funded by crypto. I don’t expect the trade-off to be 1-for-1, because the ease that fully legal, online betting provides reduces friction. But you might face a situation where, say, 70 percent of bets that would have been placed on FanDuel or DraftKings or Caesars wind up in the gray market instead with the other problems that causes.

8. Finally, before we get into the substance of the indictments, I need to remind you that these are just allegations. I’ve read several of these gambling-related indictments in the past. My personal view — an informed view, I’d like to think, but I’m not a lawyer — is that there are fewer open questions here than in, say, the case involving Shohei Ohtani’s interpreter, Ippei Mizuhara. However, the feds can be theatrical in how they present the evidence, and federal prosecutors aren’t necessarily gambling experts. The attorney Jim Trusty — good lawyer name — has questioned the allegations against his client, Terry Rozier, one of the players named in the indictments, and the NBA cleared Rozier after the league’s investigation.

The perils of player prop bets

In the sports betting indictment, there are basically two sets of allegations. One is centered around player prop bets, and particularly Rozier, who is accused of asking out of a Charlotte Hornets game on March 23, 2023 after the first quarter due to a “supposed injury” to ensure he’d hit the “under” on his projected number of points, rebounds, and other statistics.

If you’re only modestly familiar with sports betting, you might not be aware of the ubiquity of these prop bets. Traditionally, sportsbooks offered lines based on team performance — i.e., point spreads, moneylines, and over-unders (a.k.a. totals). Prop bets emerged partly because the immediate predecessor to full-blown sports betting under U.S. law, daily fantasy sports (DFS), was solely about player performance. However, in DFS, you’re competing against other players rather than the house.

9. Player props are inherently more subject to manipulation because only one player needs to be involved in rigging the outcome. In team sports, having only some players working to tank the game may not be decisive. The infamous 1919 Chicago White Sox actually won 3 of their 8 World Series games, for instance, in part because not all players were involved in the conspiracy. Likewise, a number of bets that the mob placed during the 1978-79 Boston College point-shaving scandal wound up being losers because most of the players weren’t participating.³

10. Even as someone more sympathetic to gambling than most people you’re probably reading on this story, I’m not sure I’d really care if player prop bets were banned entirely. They produce abusive behavior toward players. They also often put team and individual performance into tension with one another: you might be rooting for a guy to take a 3 so he’ll hit the over on his points prop, when passing to the cutter for an easy layup would be better for his team.

11. Furthermore, player prop bets tend to be targeted toward recreational gamblers, which is part of why the sportsbooks like them. They have lower limits, because if you’re betting hundreds of thousands of dollars on a mediocre player like Terry Rozier doing much of anything, that’s inherently somewhat suspicious. They also tend to have a larger “hold”, meaning that the sportsbook takes a larger cut of each bet.⁴ So while, in theory, a sharp bettor could probably build a good model of, say, how many rebounds individual members of the Boston Celtics were likely to obtain in a game, it’s often not worth their time because they’ll face relatively low limits, and the sportsbooks will probably restrict their accounts further if they show a propensity to win these bets.

12. In the indictment, you’ll see that quite a few different co-conspirators are involved in placing bets on Rozier, in amounts ranging from as little as $800 to as much as $107,000. Again, this likely reflects the fact that sportsbooks place lower limits on player props than on traditional bets like point spreads, so you’ll need to get a number of people involved to get the money down that you want.

13. Still, while the smaller bets probably reflect some degree of loose lips within the operation — someone’s brother’s cousin catches wind about something — the bigger bets are large in the context of player props. (If I walk up to the betting window at the Wynn with a suitcase of $100s, they’re certainly not going to let me bet $100K on the Terry Rozier rebounds under.) Some of these accounts are probably from customers that the sportsbooks have flagged as “VIPs” or “whales” because they have a history of what the sportsbooks believe to be negative-EV wagering for large amounts. The flip side of limiting the stake factors for sharp players is that the sportsbooks will let VIPs bet obscene amounts with few or no questions asked. Whale/VIP accounts are valuable because these accounts can be used as “beards” by sharp players whose action has been limited.

14. Offering the same limits to every player, or at least reducing the spread in the stakes offered, as some states are considering, would make the process fairer for all involved but less profitable for the sportsbooks. Fine, let people who have the misfortune of watching the Hornets make a $100 or $500 bet on Rozier. But when you’re letting some bettors place tens of thousands on such a bet, you’re asking for trouble.

15. As an aside, states that license sportsbooks probably ought to demand more transparency from them about how much money they’re making from different sorts of bets. I can’t find reliable estimates of what percentage of sportsbook action comes from player props. ChatGPT estimates 10-30 percent, but with low confidence, in part because these bets are often popular parts of parlays (i.e., the Chiefs beat the spread, Travis Kelce scores a touchdown, and Patrick Mahomes throws for at least 275 yards).

16. I don’t want to be too didactic here. But at a minimum, player props should probably be limited to prominent players in professional sports. Rozier is one thing — at least he was a starting point guard. But why on earth would you want to bet a substantial amount on an obscure reserve like Jontay Porter, who was permanently banned from the NBA after participating in a similar scheme to the one Rozier is accused of, unless you had inside information of some kind?

17. And while there’s some sort of protection from professional players colluding with gamblers because of their handsome salaries, that isn’t true for college athletes.⁵ As I’m writing this on a Saturday morning, for instance, FanDuel is offering bets on how many receiving yards that Yamir Knight of the SMU Mustangs will have in his game against Wake Forest. Nothing against Knight, but he isn’t even an NFL prospect, averaging just 44 yards per game so far this year.

Why are multimillionaire athletes getting involved in this?

Before we move on to the next section of the indictment, let’s pause to ask a bigger question. Rozier is not a star, but he’s made $135 million so far in his NBA career. So why are players like him (allegedly) participating in these schemes? That’s not clear, but here is an incomplete list of plausible explanations:

18. A player could think it’s easy money and/or that there’s little risk of detection. I’m of the view that professional athletes are mostly quite intelligent. But they receive exorbitant salaries at a young age, many don’t complete college, and they’re notoriously bad at managing their financial affairs.

19. They may also think they’re invincible, especially if they get the sense that the league isn’t interested in exposing their behavior.

20. They could fall in with the wrong crowd, start out thinking they’re doing a solid for their buddies, but this could spiral out of control.

21. They could be involved in other forms of (illegal) gambling and owe money as a result.

22. And/or they could be threatened by the mafia or illegal bookies, who say they will expose their gambling or other illicit or embarrassing activities.

23. And sometimes, con artistry is its own “sport”. As my podcast cohost, Maria Konnikova, has covered in her book, pulling off cons is not always rational in a narrow financial sense. People do it because it’s fun to elude detection and get away with it. And professional athletes are highly competitive. Throughout their lives, they’ve won most of the “games” they’ve played.

Inside information on NBA player availability is often reflected in betting lines

I’ve been betting the NFL this year for really the first time in my life. I’m not betting enough to make a significant profit, and while I’m off to a good start, I’d be happy to finish in the black at all because NFL lines are notoriously tough to beat. But it’s fun to see how ELWAY is doing, and I think it makes me a better modeler to see how our numbers square up against Vegas.

So no apologies for this. But honestly, the NFL has been kind of a relief, because there’s much less ambiguity about whether players will be available. Officially, both the NBA and NFL have diligent injury-reporting practices; NBA injury reports are updated once every hour, even in the middle of the night. In the NFL, though, there’s just one game a week, and players practice throughout the week. Furthermore, there’s far less tanking and “load management”. It’s rare for a quarterback to be a “game-time decision” or for him to sit the game out unexpectedly. And while quarterbacks have similar impacts on point spreads to NBA stars — Lamar Jackson versus Cooper Rush might impact the Ravens’ line by a touchdown or more — that isn’t true for non-QBs, whose availability will generally make a difference of just 1 or 2 points at the most.

24. The indictment describes several instances where information about player availability was known to insiders before the general public. For example, that “Co-Conspirator 8”, almost certainly Portland Trail Blazers coach Chauney Billups (who was also named in the poker indictment), informed his co-conspirators ahead of time that the Blazers would rest Damian Lillard and other star players in advance of their March 24, 2023 game against the Bulls because the team was tanking. I don’t know the ins and outs of how player availability information circulates to the league office, but there has to be some room to tighten up the protocols. Once you’ve made a decision on who’s playing, you ought to have an hour to report it to the league — and if you don’t, you should face fines and suspensions.

25. However, sometimes you can detect this from betting lines just because it’s so obvious that there’s private information impacting them. My records show, for instance, that I bet on the Bulls in the late afternoon on 3/24/23 and then again shortly before tipoff.⁶ The line had steadily been moving against the Blazers in a way that suggested that Lillard and other starters were highly unlikely to play, even if they were officially listed as “questionable”. Moreover, I was routinely betting against the Blazers late in the 2022-23 regular season because I suspected them of tanking both before and after they officially fell out of playoff contention. If a team is telegraphing its intentions so clearly, what does that say about the competitive integrity of the sport?

26. The 2022-23 season was also the year in which the Dallas Mavericks blatantly tanked in order to avoid having to give up their first-round draft pick to the Knicks. They were fined $750,000, but that’s a drop in the bucket for then-owner Mark Cuban — and a first-round pick is worth far more than that anyway. One of my largest bets of the season was against the Mavericks on April 7, 2023, their penultimate game of the year, in which Luka Doncic played only 13 minutes in a game the Mavs ultimately lost by three points. In the notes column on my spreadsheet, I wrote “Just a fucking weird one, what is DAL doing? Luka 1Q?” about their plan to limit their star’s action, which they’d been explicit enough about but wasn’t clear to some of the fans in the arena.

27. Although the Mavs’ game wasn’t mentioned in the indictment⁷, the Orlando Magic’s April 6, 2023 game was cited by the Feds. In that game, the Magic basically rested all their good players, and some of the co-conspirators had been tipped off in advance. I’d bet against the Magic in that game, writing in my notes “both teams trying to lose”.⁸ Much of the final third of the NBA’s schedule is compromised by these perverse incentives.

28. Another co-conspirator mentioned in the indictment is Damon Jones, an 11-year NBA player who was a close confidante of LeBron James and allegedly had access to inside information about him. Nobody is accusing LeBron himself of wrongdoing. But James was often listed as “questionable” for Lakers’ games on injury reports, which is supposed to imply roughly a 50/50 chance of playing. Betting lines often told a different story. Because he’s such a high-impact player, you could basically impute what the market thought about his actual prospects of playing. For instance, if the Lakers would typically be 10-point favorites at home against the Hornets with LeBron, but 4-point favorites without him, a spread of Lakers -5 would strongly suggest that he’d miss the game. I often bet against the Lakers in 2022-23, both because our model thought the public overrated them and because their intentions were often telegraphed by betting lines. But on Feb. 9, 2023, a game mentioned in the indictment, I was apparently too tempted by a line showing the Lakers as 2:1 underdogs at home against the Bucks and placed a small moneyline bet on them, writing in my notes “Just an incredibly weird one”. The line didn’t make any sense unless LeBron was almost assured of not playing. But, of course, he was DNP’d with a “sore left ankle” and the Lakers lost.

29. The 82-game NBA regular season is incredibly tough on players, so it’s usually easy enough to come up with some pretense of an actual injury. But as I hope these examples show, the pretense is often wafer-thin, especially when a number of players on the same team just so happen to get “injured” at once. If a player is going to be rested for “load management”, that’s fine by me, and perhaps the season should be shortened. But the league should just be more honest about why players are missing games.

30. And they should make this clearer in advance. Instances of load management should be announced at least 12 hours in advance, or 24 hours in advance if the team didn’t have a game the previous day. If there’s a new, acute injury, fine — but the league should investigate teams if this happens unusually often.

31. I’ll just leave this here as an accusation for now, since the league strongly denies it. But ESPN’s Brian Windhorst has reported that Rozier received what was essentially a paid suspension late in the 2022-23 season as the league investigated gambling-related accusations against him, but this was listed as a foot injury on official league injury reports. I don’t know who’s right here, but regardless, both the disclosure of a player’s unavailability and the reason for it should be tightened up. A chronic injury, an acute injury, a “personal matter,” and a load-management day all have different implications for a player’s availability for future games.

32. One option to reduce the impact of inside injury knowledge would be to make bets contingent on key players participating, as some sportsbooks do for baseball with starting pitchers.

And what about those rigged poker games?

While at first glance, the poker and betting indictments might seem relatively unrelated, some defendants were named in both. And that doesn’t count Billups, who was named in the poker indictment and is almost certainly “Co-Conspirator 8” in the other one.

33. Poker and sports betting, as the canonical forms of skilled gambling, are birds of a feather that flock together. At many casinos like the Bellagio, for instance, the sportsbook is located right next to the poker room, and it’s pretty common to see players sweating out an NBA game while they’re waiting for the flop.

34. The large majority of people playing poker are not cheaters, and poker games played at licensed casinos in the United States are generally quite safe. However, it’s probably also safe to assume that players who engage in underhanded tactics at sports betting are also more likely to do so at poker.

35. Although one could say that there’s a blurry line between the $20 home tournament you play with your buddies every Thursday and the high-stakes, mafia-backed games described by the indictment, I don’t think that’s actually the case. The key distinction is this: Is the person hosting the game making a profit from it? I don’t mean a shared collection for pizza, beer, a third-party dealer, or other mutual expenses. Rather, is the house taking a rake or otherwise making a profit? This is not legal advice, but home poker games without a rake are explicitly legal in many states and aren’t likely to be a target of law enforcement even where they aren’t, except perhaps for very high stakes or if they’re a nexus for other illicit activity.

36. I’ve played in plenty of home games across a wide variety of formats and stakes, but I’ve never played in a raked game outside of a casino, and I’d strongly advise against doing so. The mere fact that such a game is almost certainly illegal raises questions about the ethics of the people running it. Even if there aren’t physical forms of cheating such as rigged decks, the dealer might palm chips (take more rake than he’s supposed to). Or you might have trouble getting paid out if you win. Or opponents might be colluding against you.

37. It’s not surprising that the mafia is involved in illegal raked games. You need to fade law enforcement, facilitate the exchange of relatively large amounts of money off the books, and so forth: the mafia is good at that kind of stuff.

38. Other precautions to take when considering home games: first, you’ll want to personally know many of the people involved, especially the host. And you’ll want to have some sense of why you’ve been invited. Often, the explanations are benign (they see you as a fish — or a fun guy to hang out with even if you might be a small winner). But if you get a last-minute text from your poker buddy telling you about a juicy, high-stakes game running on the other side of town, be wary.

39. The magnitude of cheating in the alleged Billups games wasn’t very subtle. There were rigged shufflers, marked decks, and hidden communication devices: basically, every way that you can think of to cheat, these guys were (allegedly) using. But cheats are often greedy, which is actually a blessing to those of us trying to play it straight. The cheating was also so obvious in the Billups game that the poker player Matt Berkey called Billups out by name on his podcast two years ago.

40. In the podcast, Berkey said that some professional players suspected they were being cheated, but returned to the games anyway because the other players were so bad that they couldn’t imagine how they were losing so much money. But from a Bayesian standpoint, this is precisely the 180-degree wrong heuristic. The more implausible it is that you keep losing money and taking bad beats because the other players are such fish, the more likely it is that you’re being cheated.

41. Professional athletes aren’t necessarily such bad poker players. In the WSOP, for instance, I’ve actually played against some superstar athletes, including Neymar Jr. (!) and the boxer Ryan Garcia. They tend to be aggressive and not afraid of losing money, which is more than you can say for most amateurs. But the World Series of Poker is one thing, and an underground game is another. Athletes seem like fun hangs, are presumably pretty bad at poker — but also add an air of legitimacy to the games. To get an invite to play cards with someone like Billups might seem like a win-win. But gambling situations that seem too good to be true usually are.

42. As for Billups’s potential motivations, I’d refer you back to the above section on “Why are multimillionaire athletes getting involved in this?”, but add a couple of additional considerations. First, the amount of money alleged to have been scammed from players in the poker games is large compared to the Rozier prop bets: according to the indictment, “at least $7,150,000.” And second, poker is a hard game to play well. You can imagine a professional athlete who’s used to being a winner being frustrated by losing money to a bunch of nerdy tech bros and poker pros when playing in straight poker games, and eventually finding other ways to come out ahead at the end of the night.

43. Poker is a great game. I’d argue that it is very plausibly a net good for society, which I’m not sure I’d say for sports betting. It teaches discipline, analytical skills, people reading, risk management, and lots of other useful life skills. It fosters camaraderie among people from different walks of life. And it’s not actually a major profit center for casinos, typically constituting somewhere around 1 percent of overall gambling-related profits. So it’s a shame that poker so consistently makes negative headlines when it breaks into mainstream news. In general, I don’t think the community of above-board professional poker players (and “serious” amateurs, which I guess is how I’d describe myself) goes far enough in ostracizing and shaming scammers and cheaters. But the Billups games are from a different part of the poker world.

1. I’m also an advisor to Polymarket, which offers markets on sports along with many other things.

2. With that said, I wouldn’t underestimate the degree of inertia in public policy. It’s hard to make new laws, but also hard to undo them, and sports betting is also a source of revenue for the states that license it.

3. Of course, there has also long been betting on individual sports like tennis, but these sports have faced credible allegations of match-fixing in recent years.

4. For point spread bets, the hold is typically 4.55 percent; for player props, it’s usually at least 5.5 percent and goes up from there.

5. Yes, I’m aware of the presence of NIL.

6. The timing of the latter bet was unusual for me. The earlier you bet, the more profitable the lines, so I usually tried to finish up all by betting by the afternoon. While I can’t recreate my exact thought process from two years ago, the late bet suggests I was actively monitoring the situation because I suspected the Blazers might do this.

7. The problem there wasn’t inside info so much as that the Mavs were blatantly tanking and the league didn’t care.

8. While the Magic were trying to secure a better draft spot, the Cavaliers had clinched the playoffs but thought finishing with the #4 seed rather than the #3 gave them a more favorable draw.