There's No Racism in Baseball!

Well, actually it looks like there is, but less than there is in basketball.

Time magazine reports on a new study showing baseball umpires have racist tendencies in a little less than one pitch per game. And this tendency shows up in the likelihood that a pitch is called a strike when the pitcher and the ump share ethnicities, according to the paper. Under these circumstances, the chance that a pitch was called a strike increased by 0.34 percent. (This finding came after the researchers accounted for things like an umpire's tendency to call strikes, a pitcher's accuracy or control, or a batter's ability to attract balls.)

Although that might seem like a tiny amount, the authors try to argue otherwise:

It is obvious that the direct effect of racial bias on pitch calls, such as the potential for a pitcher facing a racially/ethnically unmatched umpire striking out fewer batters or giving up more walks, can alter games, especially close ones. The indirect effect--when players anticipate the effect of a biased umpire and strategically alter their behavior--may, however, have an even larger impact on outcomes.

Frankly, I think that's a HUGE stretch. The fact that going purely by the numbers shows a very tiny bias among umpires leads the researchers to try to strengthen their case by calling on all sorts of reasons for why racism is much more significant in reality.

But as many economists like to say, the dismal science is all about incentives. In this case, let's look at the incentives of the researchers: they just spent their valuable time putting together this paper, so on a personal level, it might not be self-affirming to report racism might not be a huge problem.

How can I prove this? I probably can't, just the same way the researchers don't prove that the underlying effects of the one-pitch-per-game bias really changes the outcome of games.

Is there racism among umpires in baseball? Almost definitely. (Where isn't there in America?)

Does it effect who wins a ballgame? This study suggests it doesn't. The researchers show umpires' racist tendencies are reduced when their decisions are under greater scrutiny: when there's a full count, when game is played in a QuesTec-equipped stadium, and when crowds are large.

The study was inspired by research claiming N.B.A. referees are similarly biased against players of different ethnicities.

But the N.B.A. study found stronger results: Players got 4 percent fewer fouls and scored up to 2.5 percent more on nights when their ethnicities matched that of the officiating crew.

What could be more interesting is if the researchers could find out which umpires might be more racist than others. Unfortunately, because of the way the study was constructed that was not possible, Daniel Hamermesh, a professor of economics at the University of Texas at Austin and one of the authors of the study, told me.

Meanwhile, I really liked some of the stats the study was able to dig up. Here are some graphs for the baseball geeks:

Related:

• F.A.Q. by the Study's Authors (pdf)


• Sabermetric Research Blog's Take on the Study


• The Malleability of Automatic Stereotypes and Prejudice