What Do NBA Referees and MBA Teachers Have in Common?

Over at Marginal Revolution, Tyler Cowen draws an intriguing parallel between accusations made by disgraced NBA ex-referee Tim Donaghy, and models of collusion.

While David Stern has denied explicit collusion between the league and the refs to influence game outcomes, Tyler argues that there may instead be implicit collusion: refs may simply perceive that the league wants them to produce certain outcomes, and the refs will deliver unless the league takes a strong contrary stand. This may explain why star players are so widely perceived to get better treatment, and it could also explain why some of Donaghy’s accusations resonate.

And while strong financial incentives often lead to better job performance, in this case they actually exacerbate the problem, as they make it even more important to the refs that they deliver the outcomes they believe their bosses want. It is an interesting and subtle argument: Read it here.

My own introspection — based on MBA teaching, rather than NBA refereeing — tells me that Tyler’s argument is quite reasonable. Officially, Wharton tells me that my job is to teach students important truths, but sometimes I fear that really my job is to entertain the students instead. You can see that my dilemma is a lot like that facing the NBA refs. And while I’m ashamed to admit it, my teaching does sometimes reflect the entertainment, rather than truth-seeking objectives.

Moreover, I acted more like this when the incentives were highest (when un-tenured) than I do today. If Ivy League profs act this way, is it so crazy to suspect that NBA refs occasionally respond to the incentive to create entertainment, even if it clashes with truth-telling?

David Stern has sought to reassure the public that all is well and that Donaghy’s accusations should be dismissed out of hand. But the fact that the league often issues blanket denials — sometimes of limited credibility — is a real problem for them here. Firms sometimes find it beneficial to deny accusations that are true (like this one), but the long run cost is that they have less credibility when denying damaging falsehoods. Unfortunately for the NBA, fans have no way of discerning the true denials from the PR cover-ups. Moreover, if Tyler is right, it isn’t enough to simply show that the NBA did not ask the refs for particular favors, they need to show that they actively (and equally) discouraged all refereeing errors. If such evidence exists, I would strongly urge Commissioner Stern to make it public.

I discussed these issues at length with ESPN’s Henry Abbott, and he wrote up our conversation for the True Hoop blog.

More generally, how should the League respond to gambling-related concerns? I’ve argued before that the problem with current U.S. gambling laws is the disjunction between our formal laws, which are among the strictest in the world, and the law-in-action: U.S. gamblers have been pushed offshore, and so, in effect are governed by the most lax gambling regulation in the world. Like Rep. Barney Frank, I believe that a well-regulated gambling sector may be better than the current Wild West approach.

Related reading: More on my approach to gambling laws in this New York Times Op-Ed, written when the scandal first broke (see Dubner’s $0.02, here); Joe Drape on how legalizing gambling may make it easier to catch bad guys; and Brian Goff at the Sports Economist expands what it means to manipulate game outcomes.

And while this post has been about how economic analysis can help us understand sports, Andrew Leigh has a nice post on how sports can help us better understand economics.

Update: The WSJ‘s “Numbers Guy”, Carl Bialik, critically assesses the claim that there have been too many 7 game playoff series, here. It’s a clever rejoinder to some sloppy headlines.


I think NBA refs are paid per series in the playoffs, but I don't have a reference to give you for that.

Anyway, refs have an incentive to make the league (their employer) happy. Refs are graded by the league and the ones with the best grades (probably with some seniority considerations) get the playoff series. On the other end of the scale, refs who do badly by the league's criteria could be out of a job. So, the refs try to figure out the league's performance criteria (some of which are not officially spelled out) and ref to them.


For the playoffs, NBA refs are paid the same way the players are: they split an amount of money for every round they work. If the series goes 4 games or 7 games, they get paid the same amount. On the other hand, the NBA decides what refs get to work the postseason based on their performance. If referees feel that drawing a series out so that the NBA makes more money helps their chances of getting picked next year, they may be responding to an indirect financial incentive.

Mike B

In theory, if "high minded" policy makers truly want to put the kibosh on sports gambling, they should welcome betting scandals instead preaching against them! Why? Because if sporting events are widely seen to be corrupt and rigged, the betting public will stop betting on them.

The only people who really have anything at stake in these scandals are the gamblers themselves! It makes no sense for the anti-gambling government to go out of their way to make sports a safe venue for gambling. If sports are a form of entertainment, then it does matter if they are rigged or not behind the scenes. Look how popular professional wrestling is and that is blatantly rigged. Only when gambling is involved do sports become something other than entertainment and rigging becomes corruption.

If the emotion driven policy makers truly wanted to eliminate the corrupting influence of gambling the only solution would be to run all professional sports like the WWE.



During the 2008 NBA regular season the home team won 60% of the time, 739 out of 1230 games. During the playoffs the home team has has won more than 75% of the time. Overall this season the home team's winning percentage is 61%.

Why? Do you really believe that pro basketball players play better when the most of the fans root for them and worse when the fans are against them? Or is it the officiating?

If the fans had *any* effect on the play, it should show up during free throw shooting. When the visitors go to the foul line, the hometown fans behind the basket scream, jump up and down, and wave banners and other objects in an effort to distract the shooter.

I decided to check the efficacy of this behavior in a sample of the games. I decided -- ahead of time -- to use the statistics from the third and fourth games of each of the 15 series (8 first-round, 4 quarter- and 2 semifinal series, and 1 final series), a total of 30 games. Since the teams always change courts after the third game, in this sample each team is home once and away once in every series.

Here are the results: The visitors made 573 out of 764 attempts, or 75% The home team players made 618 out of 828, or 74.6%. There is evidently no home court advantage in foul shooting.

The difference in the number of fouls called, about 2 per game, favors the home team but is not enough to explain the outcome by itself. Where the officials can most readily affect the game without showing up in the stats is in the fouls they DON'T call.