Thursday, August 16, 2007

Analyzing MLB Umpires and Race

There's an intriguing study by a group led by Daniel Hamermesh at the University of Texas at Austin about whether there may be a correlation between baseball umpires calling strikes and whether the pitcher is of the same race or ethnicity as the home plate umpire. According to news reports, the study reviewed three MLB seasons (2004 through 2006) and they believe there is a positive correlation between the strikes called and the race or ethnicity of the pitcher being the same as the home plate umpire.

It is great that journalists are commenting on the performance of referees. It's too bad it takes a team led by an economist to do the hard analysis rather than journalists themselves doing the work. Why are journalists not writing insightful pieces of umpire or referee performance? Are they afraid of being shunned by the sports leagues? Are they unwilling to do the hard work of crunching the statistics?

It's hard to tell from the coverage of the study, but it seems great that the study's authors may be focused not on intentional racism, but just on the possibility there is some unconscious, unintentionally difference in performance based on race. That would be great because it would emphasize that nobody is accusing the umpires or referees of intentionally trying to affect the results. It is just that umpires are human beings and can have unconscious tendencies, despite their honest efforts to be as neutral as possible. And that's one of the keys to analyzing referees -- just because you may see some variation in performance does not mean you are accusing people of acting illegally. It can be an honest, unconscious effect.

If it is an unconscious effect, then it is great for people to analyze it and point it out so that people can try to address unconscious tendencies. According to news reports, the umpires do not show as much of a small tendency in big games, important pitches, or where there is a huge crowd. So scrutiny might help improve umpires' performance. Which is even more reason for people to analyze umpires and referees -- more scrutiny may actually improve their performance!

There are, of course, many reasons to wonder about the significance or accuracy of the study, which I can't figure out because I haven't analyzed the study itself. Questions include whether the sample size is large enough, how you know whether the pitch truly should have been called a strike, whether a deviance on 1% of the pitches really is significant, and whether it controls for factors such as being more lenient with handsome, popular, or successful pitchers. There's also a question of whether a correlation necessarily means causation.

Regardless of whether the study's findings stand up over time or are considered statistically significant, it is great that they are studying referee and umpire performance and that they are guarded in drawing conclusions.

It's a sad fact that sports journalists are not doing any of the analysis and we have to rely on people completely outside the large hordes of people paid to write about sports to give us some intelligent analysis on how referees and umpires do. Of course you won't hear much of this from players, coaches, or owners -- because many sports leagues have a gag rule and will fine or suspend people who comment on referees and umpires.