NBA Racism?

A Professor at Penn and a Grad Student at Cornell have released a study showing that White NBA refs call fouls at a higher rate on black players than whites, and vice versa. The white refs have ahigher rate of correlation. They conclude that if talent were the same, a team could pick up a few games a season by tailoring its team's racial composition to the referees. Commissioner David Stern did his own study, finding no problem, but the New York Times showed both studies to a group of academics, who unanimously prefferred the Penn study.

I have not read the study, but by the end of the day I will.

NBA players who have also not read the study are calling it ridiculous, using anecdotal evidence in support of a race-free NBA and resorting to cliches like "statistics can say anything you want them to". This is asinine. The study in all likelihood uses a technique called statistical regression, which determines correlation by removing outside factors, and was the primary basis for the book Freakonomics.

No one is saying that race is acting on a conscious level.

However, David Stern owes every fan this: do the study again, and do it right. You have the information and the money, and some statistics are only available to you. (Who gets disciplined, who made which call). Sort this out, David. Do it now.


While you're at it, here are a few ideas:
Get all of the facts. Take about ten years of games, know who called which foul, and then take it a step further. Isolate for all the factors. I suggest each player have the following categories: race, position, nationality, years in the league, and all-star berths. The last one would be a boolean variable (yes/no) and once a player has been an all-star, then they get yes for the rest of their carreers. Next, eliminate the fouls that don't count. Intentional fouls at the end of the game can have no racial motivation, and neither can fouls after which a player self-identifies. Finally, know the refs' age, race, sex, experience, and marital status. Now on each foul there is a matrix working three ways: The player, the refs, and his or her peers.

Now you can use regression software available on every campus in America. You can find the individual tendencies of every ref, and find out overall trends too. Is it race? how about nationality? Are white refs even friendlier to white americans? What about when you exept centers, who traditionally get more fouls? Now try it for starters, since they're less likely to hack-a-shaq. Do superstars get star treatment, and is that exacerbated by years in the league?

Until that report comes out, the NBA is turning a blind eye, and are complicit in whatever bias the refs have.