"Study of N.B.A. Sees Racial Bias in Calling Fouls"

sweet! break out the krylon!

 

Spraypaint Chuck white and turn him loose!!

 

I think Reggie already Retired.

 

What's really interesting here is the reaction to the headline. The league came out with some quick denial work, throwing out the 'those guys don't have the resources we have' defense and saying they have different conclusions. Of course they will refuse to publish their conclusions and pray this thing gets swept under the rug.

Then there are the 'journalists' who believe this paper isn't legit without having read it and without knowing the math behind it. I've been hearing these guys on local talk radio and they have quickly morphed the paper into something it's not, it's kind of embarrasing that that these guys won't do their homework.

 

Just heard Mike Tirico (a black sportscaster for ESPN, btw) and had a few comments on this:

1. The study was based strictly on box scores, not video. Thus there was no direct accounting of the race of the ref for each call.

2. The headline is shocking, but w/ only a 4% difference it's a more provocative and slightly misleading premise.

3. Plenty of black stars get a pass. Superstars get a pass period. Jordan, Kobe, Wade, Magic, etc always had more calls go their way then not.

Now I only skimmed through this article so I have no idea if Tirico is correct (judging by the responses and what I skimmed there seems to be direct accounting of the race of the ref for example). Just putting out another person's opinion on the matter from one of the few sportscasters on the NBA who I think is pretty solid.

I think there's certainly a place for this type of research. Whether by the NBA, an independent agency, even a team owner like Cuban. And I'm not ready to rule out the possibility that there could be more biased calls based on race. But I'm a little bothered that this seems to be some grad student's thesis. It's like this student was running out of time and ideas and wanted to take a shortcut. Quite frankly, doing this type of 'research' is only time consuming and doesn't really require much else. Maybe use of some undergrad level statistics in his analysis. Using this as an example of work-place discrimination doesn't hold much water, considering the NBA is a totally unique employer that has few parallels to any other 'workplace.' Especially when this only uses the basis of 'calling fouls'. Seems to reflect poorly on his professor and student that they thought this was a worthwhile study based on his coursework.

 

Check this link out:

http://www.freakonomics.com/blog/2007/05/02/racial-bias-in-nba-refereeing/

It's a comment by Dr. Levitt, the author of Freakonomics. It's a great book.

 

Has anyone brought up the fact that in the last two minutes of a close game, when teams are purposely fouling and you'll hear more whistles than in the previous 18 minutes of the contest combined, ten out of ten players on the floor are usually black? Doesn't this skew anything?

 

My bad. They "opened the floor for discussion about their findings."

 

Was this really Page 1 of the NYT?

 

I've got a feeling that if Yao was an American citizen, the screwing would be a bit less than it is.

 

Everybody pull up a chair and read:

http://insider.espn.go.com/espn/blog/index?entryID=2855205&name=hollinger_john

 

Of course, but what conflict can be derived from sensible thought? Just shut up.

 

but that doesn't matter. as someone else mentioned, the variable is the refs, and more specifically their race, not what the players do.

if white guys barely commit any fouls but both white and black refs both call them for 1 foul per 48 minutes, while black guys commit a lot and both white and black refs call them for 5 fouls per 48 minutes, there would be no bias b/c the white refs would've reffed the black players just like the black refs (and same for white players).

what they seem to be saying is black refs call black players for X fouls per 48, and white refs call them for 1.04X fouls per 48 minutes. and then the reverse effect, though smaller, for white players.

whether it's all actually true or if looking at the specific ref that made the call would change things, i don't know, but that's what they're saying.

 

this could very well be true.

a study must be conducted to determine the racial composition, however.

 

ridicolous and inflammatory study to broadcast.

Absolutely not necessary by any stretch.

who funds this research?

 

Exactly. I'd be curious if they looked at this stuff as panel data - has it been getting worse or better? It wouldn't be that hard to do using a fixed effects model.

I think the title is meant to spark controversy, but the results are pretty legit.

 

Ok, thanks for the heads-up, but does it take into account the obvious foul that three refs (two white, one black) call on Stephon Marbury fouling Jameer Nelson in the backcourt with 40 seconds left in the game? I'm not questioning, I just haven't made it through the entire UPenn document.

 

And once again, the effect (regardless of sample size) is so close to zero, that it just looks as if these guys are in it to get the study into the papers. Hyperbole like that "“Basically, it suggests that if you spray-painted one of your starters white, you’d win a few more games,” Mr. Wolfers said." bit seem like a ploy to get his name out there.

From "Ajax" at the APBR forum:

"However, I think there is a confusion of significance and effect size. The effect is 0.16 +/- 0.04 fouls per 48 minutes assuming a completely white crew compared to a completely black one. If a starter plays 36 minutes, we talking about one extra foul per 8.5 games for a black player."

Two fouls in 17 games? How does that change a team's ability to win?

 

Again, what is the relevence? It would only be relevant if only white refs called that foul.

It's amazing how posts there are in this thread by people who either didn't read the article or completely misinterpreted it.

 

Agreed that it's overstated, but I think you may be understating it. If I understand these effects right, that .16 fouls per 48 minutes is per player. A hypothetical, stylized team that plays all black players would then have 5*.16 = .8 incremental fouls called per game, or 4 fouls more every 5 games. Now, does that matter, in terms of winning and losing? I think it could mean a game or two over the course of an 82 game season, but I doubt anything much more than that. To put that in perspective, the range of fouls per game by teams go from 19.36 to 25.20 (ironically, Utah). The median is around 22.28; .8 fouls per game would represent around 3.5% that. Does that matter in wins and losses? There's not really an exact answer.

*IF* the effect were more pronounced in the playoffs, or in close/late game situations, there might be a better case for this mattering. As of now, it's an exercise in academia with no real policy implications or insights.

IMO, the "racial effect" is there and real (ie, statistically distinguishable from zero, not chance), but does not materially affect winning and losing.