Racist Referees

Alan Schwarz writes a column about a paper by Justin Wolfers and Joseph Smith in today’s New York Times. The paper Racial Discrimination and NBA Referees looks at how referees call fouls against players of different races. The results are very interesting.

An academic study of the National Basketball Association, whose playoffs continue tonight, suggests that a racial bias found in other parts of American society has existed on the basketball court as well.

A coming paper by a University of Pennsylvania professor and a Cornell University graduate student says that, during the 13 seasons from 1991 through 2004, white referees called fouls at a greater rate against black players than against white players.

Justin Wolfers, an assistant professor of business and public policy at the Wharton School, and Joseph Price, a Cornell graduate student in economics, found a corresponding bias in which black officials called fouls more frequently against white players, though that tendency was not as strong. They went on to claim that the different rates at which fouls are called “is large enough that the probability of a team winning is noticeably affected by the racial composition of the refereeing crew assigned to the game.”

You may recall Wolfers from his work on point shaving in NCAA basketball. Similarly, Wolfers has found a way to use sports as a laboratory to examine human behavior in a controlled setting. When I see a study like this, I slap myself in the forehead and say “why didn’t I think of that?” I’m jealous. What an interesting way to examine those subtle racial biases that are so difficult to study in the general population.

Studies of racism in sports are very common. Why? Because one of the hardest things to control for in studies of racial discrimination is performance. You might look at a sample of workers who get promoted and who doesn’t, but there is always the possibility that other factors correlated with race influence the promotion decision. Controlling for this is much less difficult in sports because we can control for performance characteristics.

Using Dave Berri, Stacey Brook, and Martin Schmidt’s “Win Score” measure, the authors find that the effect is big enough to affect a few wins in a season. As Wolfers puts it “Basically, it suggests that if you spray-painted one of your starters white, you’d win a few more games,” Mr. Wolfers said.

It’s not surprising that the NBA disputes these findings with data that it will not share. I think there are several legitimate reasons to keep this data private, But, the league also has a duty to take the Wolfers-Smith result seriously and to investigate this issue further. In no way do I think any NBA referee consciously uses race to referee a game. But in a fast-paced contest, where decisions must be made immediately, it is not surprising to see such bias sneak out. As this story plays out, I hope that we will learn more about race and society.

12 Responses “Racist Referees”

  1. Brian says:

    JC, one comment that I heard somewhere this morning asked a question that I thought was a good one (I will offer the disclaimer that I have have not read the study). Do we know for sure that these statistics prove that white referees are calling more fouls against African-American players? Could they also demonstrate that African-American referees are treating A-A players favorably by calling fewer fouls on them?

  2. JC says:

    I don’t know. This morning was the first time I saw the paper, so I haven’t read it yet. As with any paper, there are tons of potential problems. However, given the sensitive subject matter, you better believe it’s going to get a serious vetting. I will say that the three experts who gave a vote of confidence to the paper–Ian Ayres, Dave Berri, and Larry Katz—are all highly respected researchers.

  3. Doug says:

    As is the tradition, I will spend 5 words praising the study, and then a few paragraphs on “yeah, but it’d be great if they could look at this and this and this other thing too!”

    This is a fascinating study!

    OK then, now that we’ve got that out of the way…

    A couple of quick thoughts.

    1. It would be interesting, but presumably not possible with just boxscore data, to break this down also according to the race of the foulee. I.e. black on black, white on white, white on black, black on white.

    2. It would be interesting to see what the rates look like for all these International players who have come into the league in recent years.

    3. Their data goes back a long way. I wonder how things look if you consider only, say, the last five years.

    Well, OK, let me praise the study a bit more, or at least counter one potential criticism of it (the one the NBA is apparently hanging its hat on). One big criticism of this study is that it doesn’t take into account the identitiy of the referee who actually blew the whistle. It only takes into account the general composition of the crew (3W, 2W1B, 1W2B, or 3B).

    Assuming they have appropriately controlled for all the relevant factors (which I can’t make a judgement on until I’ve seen the paper), this study finds a significant tendency for, e.g., a 2W1B crew to call more fouls against blacks than a 1W2B crew. The natural conclusion, then, is that white officials are slightly more likely to call fouls against black players than against white players.

    It seems to me that, if that’s not the case, then the only way they could have gotten the results they did was if white refs were more likely to call fouls against whites when their officiating partners were black than when their officiating partners were white. That would be an equally interesting result, if not moreso.

  4. Gordon says:

    Brian: the results actually seem to show that white refs call slightly fewer fouls on white players (or, you could say black refs call more), while ref’s race plays essentially no role in fouls called on black players.

    I have to say that the magnitude of the impact on team wins seems much smaller than the authors claim. They focus on comparing a 0-white ref squad to a 3-white ref squad, the extremes. But let’s say the ‘correct’ percentage is 17% white refs (same as for players), a reduction of 51%. If we fired all those white refs and hired Black refs, that would mean an increase of about .09 fouls per game for each “additional” white player a team had, compared to its opponents. And let’s say there is a team that has 20% (1 player) more playing time for white players than average (a huge difference). So that team will now commit 7 more fouls per season, a cost of about 3-4 points. I forget what the point/win converter is for NBA, but it would take many seasons for this to translate into a singe additional win.

    Now, the authors also say there is an impact on points, rebounds, etc., though it’s hard to tell how much, if at all, this involves “double counting” the impact of fouls. So there may be a more significant total impact on game outcomes. But in terms of fouls, the impact seems very small.

  5. Jason says:

    Exactly how many white players are there in the NBA? People talk all the time about how “there are no blacks in MLB anymore” yet I cannot name a single white NBA player who was born in the USA and the only white guys I can think of who play there are all foreign born. Does anyone who actually knows and cares about the NBA have any idea if the number of white players is so small that no significant conclusions can be drawn from this because the sample size isn’t large enough?

  6. Gordon says:

    Doug: They don’t have player-on-player data. But it’s interesting that it is white players, not blacks, whose foul rate seems most impacted by refs’ race. Over 80% of white players’ fouls (we assume) will involve a different-race matchup, while less than 20% of blacks’ fouls will. So perhaps it is mainly when a black fouls a white, or vice-versa, that elicits differing reactions.

    Also, the results are mainly driven by the all-white and all-black ref squads (the latter being quite rare). Almost no difference between 2/1 and 1/2 squads. So there may also be a group solidarity factor at work. In which case, just trying to ensure there aren’t many single-race ref squads will fix the problem (to the extent there is one).

    Table 3 has a good summary of a lot of this data.

    Jason: whites have 17% of PT, according to the study.

  7. Kevin says:

    I dont have the exact number but the majority of white players in the NBA are very small. As a fan who watches the NBA and basketball, I find the study interesting but dont think it means anything. Most “white players” are role players that rarely play, or tall euros who shoot three’s all day. Also the majority of the “white players” dont play defense very well and their opportunites to foul or get a foul called on them is very small.

    Id rather see an in depth study that focuses star vs. rookie foul treatment, as well as seeing if a foul is legitimate or not. I think if a foul is a foul, race doesnt matter, but if blacks get whistled for more illegimate fouls, or vice versa, you then would have a real intersting study. I think this is a start, but you have to dig real deep to get a truer answer to this racial discussion.

  8. KevinH says:

    I wonder what the effect of this study might be on officiating. You have to think that the refs are going to start second-guessing themselves, wondering “did I just blow that whistle because Kobe Bryant is black?”

    Referees (and anyone who has a job) need to spend some time in self-examination, making sure that they haven’t become tangled up in prejudice. But that should be done privately, and then referees can focus on what really matters out on the basketball court: calling a fair game.

  9. Nolan says:

    I really don’t think there is a legitimate reason for the NBA to keep its data private. They could easily hand over the raw numbers without betraying who the particular refs are. By the sound of it, the NBA’s study is very weak.

    Anyway, the NBA has overreacted – this isn’t an attack on the NBA or the way it trains its officials. It isn’t an attack on anything, in fact. It’s an observation and, if the NBA had any sense, it would simply acknowledge that and pledge to continue to give its officials world-class training.

  10. Adrian says:

    Hey CJ, I’ve been coming to your blog for about 2 1/2 years now and I love the book. I take it this question has not been studied in baseball. Also, in the article the authors discovered that “veteran players and All-Stars tended to draw foul calls at different rates than rookies and non-stars.” This seems to confirm what many fans have suspected. Are their any studies that confirm or debunk the proposition that good pitchers get an enlarged strike zone? I found you chapter on managers lobbying for increased strike zones very interesting.

    Thanks,
    Adrian

  11. Heather says:

    yet I cannot name a single white NBA player who was born in the USA

    Off-hand, I can tell you Kirk Heinrich, Nick Collison, and Raef LaFrentz were born and raised in Iowa (and then all played for Kansas, the traitorous bastards). Paul Shirley. Scot Pollard. Fred Hoiberg (born and raised in Iowa, played for Iowa State) was in the league while the data was collected. And I don’t even follow the NBA!

    But I can’t draw a conclusion of race being a factor in foul-calling unless the game footage was studied. What percentage was a legitimate call? How many legitimate fouls were uncalled? Do black players get called for more fouls because they commit more fouls? I don’t think it’s a far leap to wonder if the culture the player was raised in affects their style of play, because I’ve been watching it at Iowa State for years–the kids who spend hours alone on a farm shooting threes don’t play the same as the streetballers recruited from the Bronx.

  12. Hobbes says:

    Is it possible that players change their behaviour slightly depending on the racial mix of the referees?
    NB: I haven’t read the study, this may have been addressed by the authors.