Wolfers’ Study Full Of Holes, Questions And Suspect Hypothetical Calculations

By Gregory Moore
Updated: May 7, 2007

NBA SAN ANTONIO — Does race color our evaluation of others? We provide new evidence on racial biases in evaluation, by examining how the number of fouls awarded against black and white NBA players varies with the racial composition of the refereeing crew. This sentence from the “Racial Discrimination Among NBA Referees” written by Joseph Price and Justin Wolfers. The study has been out for a few days now and when I went through it, I have to admit that I was more than just a little befuddled why a couple of scholarly collegians would even attempt to try and justify the results that they came up with.

If this study had to pass a true sniff test, it would have flunked before it even made the first revision of May 1, 2007. Let me be frank about this study. Mr. Price and Mr. Wolfers have done the fans, the players, the teams, management and the NBA disservice with this study.

Call it a sham. Call it worthless. More importantly, let’s call it what it truly is; a waste of money and effort on a subject that is so minute, so inconsequential, that there is no true reasoning of justification. Or is there?

As I mentioned, I was very much discombobulated by this study. To give some of you a little background, I am not some Internet columnist with an axe to grind. I have been covering the NBA since 1993. I am proud to call one of the pioneering African American referees, Ken Hudson, a friend. I am glad that in my almost 15 years of seeing some of the greatest athletes in the world run up and down the court and write or talk on a radio station how this player did this or that for his team.

I can unequivocally pronounce that in my 14 years of being a member of the NBA and the San Antonio Spurs that I have not seen one bit of racial biasness by any referee in any game that I have witnessed first hand. To say that the NBA referees who are white have a grudge against the Black NBA players lets me know exactly where this study is coming from.

What the Price/Wolfers study shows this NBA veteran is nothing more than two guys who were able to garner some federal or outside funds, and try to say the NBA is racist. That’s a little hard to prove. According to the 2005-06 Season Racial and Gender Report Card for the NBA written by Dr. Richard Lapchick, “The NBA has had the top grade among the men’s leagues for race for all 12 previous reports over nearly two decades.”

The Lapchick study is as non-biased as you can get when it comes identifying what league has given African Americans a ‘fair’ shake in the sports world. The NBA’s player association is ran by an African American. The study points out that in the NBA, almost 78% of the players were people of color.

The percentage of African-American players decreased to 73% since the last Report Card and was the lowest percentage of African-Americans since the ’90-91 season when it was 72%. The percentage of Latino and international players continue to increase and that as of the end of the ’05-06 season, 62% of the NBA’s referees were white, 34% were African-American, and three percent were Latino. One of the 61 referees was a woman.

This information is crucial because in order for this recent study to have truly in credibility, there has to be a systematic presence of racial biasness by the league. That premise is very much hard to prove based on one following principle: the league’s players are made up of a majority of African American players and that make up has been that way for well over two decades.

While supporters of this study will say that is the very premise, I would challenge anyone to realize that there have not been any such legal cases of racial discrimination by a player saying that a referee has discriminated him against. You also have to look at the working relationship of the two parties involved. The mere fact that the referee crew is a majority Caucasian group has no bearing on the calls that are made in a game; experience is that factor.

Of the 61 referees in the league, the majority of the officials have over a decade of experience and that includes several African American officials. Also, you have to question whether the researchers have personally witnessed any of the games they use in their report or whether this was just statistical data that was collected.

TRYING TO DEBUNK WHAT THIS STUDY SAYS IS TRICKY To debunk this study, from my own perspective, I had to do something that Wolfers and Price didn’t do; get league information on referee performance. That was not an easy task in the beginning but with the help of very league that is being criticized and/or studied. By Friday morning, the NBA’s Joe Litvin, president of league and basketball operations, made available to me a summary of their own study as conducted by the Segal Company.

“We reviewed over 155,000 individual referee calls in 3,482 games from November 2004 through January 2007″, the summary says. “Within this group, there were about 148,000 relevant calls, since they involved a player and an official who were either white or Black. Calls involving officials and/or players who are neither black nor white were excluded.”

When looking back at the Wolfers/Price study, they made an important charge as to why they felt that race was a contributing factor of the calls being made against certain players.

“The evaluators we study — NBA referees — are effectively randomly assigned to each game. Moreover, the number of games played is large, so we can assess both a very clear baseline rate at which individual players commit fouls, and also a clear baseline for the number of fouls called by different referees. Against this baseline, we find systematic evidence of an own-race bias.”

“Players earn up to four percent fewer fouls or score up to two and a half percent more points when they are the recipients of a positive own-race bias, rather than a negative opposite-race effect. Player statistics that one might think are unaffected by referee behavior are uncorrelated with referee race. The bias in refereeing 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.”

That paragraph in itself is very bias and is based upon outside literature that does not have any bearing on the work environment within the league. What this study has done was to heavily rely on the experiences, studies and writings based upon what I like to call “real world” applications and try to apply them to a sports environment that is nothing close to what they are referencing.

For example, the Wolfers/Price study refers to a study written by Antonovics, Kate and Brian Knight entitled “A Look at Racial Profiling: Evidence from the Boston Police Department”. If this study is trying to imply that there is a racial discombobulation between a Black player and a White referee as a sign of the Black player not respecting authority, this paper would indeed be very well off base. In the 15 years that I have covered the league, I have only heard of maybe two times when a player physically went after a referee for making a bad call.

What is also troubling is the fact that this study has went to great lengths in using statistics that are very erroneous in their evaluation of whether a black player gets more fouls called against him than other players. To debunk this study on that premise, I looked at the study that Dr. Lapchick did back in 2006 on the NBA’s racial diversity.

In his study, there were 315 African Americans who were players in the league. The total number of players in the league, at that time of his report was 512 players. African Americans made up of 73% of the league last season. What also hurts the Wolfers/Price study is that if there are only 61 referees in the league and 34% of them are African American that means that 64% of the remaining referees are white. That percentage equates to 40 Caucasian referees. The ratio of a white referee having interaction with a Black player is almost an 8:1 ratio.

That high ratio number is significant because the Wolfers/Price study wants to suggest that this is the very reason why there is racial biasness amongst referees. However what this suggests is the simple fact that in a field where competition is and job slots are few, it is difficult for a person of color to break into the officiating ranks in the NBA solely based upon who is still working and who has left.

According to the NBA Officials register, a document that Mr. Wolfers and Mr. Price were not able to procure because it is not readily made available to the public, there are actually 65 referees in the league and that included three rookies in the 2005-06 season. And here is something else that this study fails to inform the reader; two of the top supervisors of the referee cadre are African American; Stu Jackson who is the senior VP of Basketball Operations and Ronnie Nunn, who is the Director of Officials.

DEBUNKING STUDY USING REAL GAME DATA When it comes to actually tracking whether certain referees are biased towards certain players, again the league’s own study helps debunk Wolfers/Price by using the criteria of that paper. For example, the Wolfers/Price study has a table that suggests racial bias. However if one were to closely examine the chart in this table, one can see that the researchers used some variables that do not have any bearing on an outcome of a game.

Variables such as height, player skill position and whether the player was an all star or not DO NOT having any bearing on the outcome of a game. To bring that hypothesis into play, what I did was pick a random game from this week and analyze the box score and play by play of the same game.

For my example, I chose Sunday’s Western Conference Semifinal between the San Antonio Spurs and Phoenix Suns. The Spurs won 111-106. In the game, 49 fouls were committed between both teams (San Antonio committed 25 personal fouls while Phoenix committed 24).

Of the fouls committed, The Suns’ starting five committed seventeen of their 24 fouls with Raja Bell committing the maximum six fouls allowed per contest. For the Spurs, their starting five committed thirteen of their twenty-five with Bruce Bowen having four fouls. The referee crew for the game was James Capers, Bob Delaney, and Mark Wunderlich.

For the record, Capers was the only African American referee on the floor however he was the referee with the least amount of officiating experience at ten seasons in the league. Delaney is in his 19th season and Wunderlich is in his 16th season.

What the Wolfers/Price paper wants to portray is that because of the lack of Black referees in the league, the calls would be very much biased. However, because we are not using a ‘live’ game that has been witnessed and not some statistical model of information that has been archived, I can say without any hesitation that their study is very much flawed.

In order for their study to be accurate, Wolfers and Price would have to show that certain players who are Black are not given a fair shake during a game. However if you take a look at the box score for this game (http://www.nba.com/games/20070506/SASPHX/boxscore.html), you can readily see how this would not be the case.

The stars for both teams were able to play their games efficiently and not be hampered by a bad call that reflected the game in an adverse manner. For example, Amare Stoudamire scored 20 points, had 18 rebounds and played 38 minutes despite having four fouls.

Stoudamire is considered one of the team’s best players. The same can be said for Tony Parker. Parker had two fouls but scored 32 points and had eight assists in 37 minutes of play. In both cases, neither player was hampered by what the paper would suggest as biased calls.

This fact can also be bolstered by the point that during the game, there were several instances in which a white referee called the foul on a Black player who just happened to foul another Black player.

One such play happened at the 10:32 mark of the fourth quarter in which Parker drove to the basket against the Suns’ Shawn Marion with the shot clock running down. Parker hit the lay up and was fouled by Marion. The referee who made the call was either Delaney or Wunderlich as Capers was away from the call.

STUDY NEEEDS MORE WORK TO BE TAKEN SERIOUSLY The conclusion of my own little study as to whether the Wolfers/Price study is correct favors that of the league’s own assessment. Their study was faulted based upon the use of data that went back to the early 1990s, a time when the league did not have the measures in place they do now for evaluation of each call.

Although referees were evaluated back during that time frame, from at least 2000 forward, with the ability to have high speed internet, the league has been able to have the best officials in the world. Wolfers/Price fail to take into account any CURRENT information that would have been negative to their cause.

If their research was truly to be inclusive, they would have used not only the box scores and some information from NBA.com, but also look at the play-by-play sheets of the games sampled and have access to the evaluations of the referees involved. That last part of their analysis is impossible because what they or I cannot do is get access to this very important piece of information.

However, what the researchers could have done was ‘google’ the most recent seasons of games and looked to see whether their hypothesis held true. According to their own report, they failed to do so.

Even though my own research was not as in depth or as ‘scientific’ as the Wolfers/Price or Segal company studies, I think that I have been able to prove that the Wolfers/Price project was indeed a flawed exercise because they did not have the expertise of reading the box scores, play by play sheets and knowledge of covering the sport to make an educated diagnosis of their own questioning.

The league’s referees are not racist nor can the Wolfers/Price study prove this assertion beyond a shadow of the doubt. That premise is proven by the use of any box score and play-by-play sheet for any game final you can pull off of NBA.com’s website.

The study also is debunked by the mere fact that those who have access to league materials that allow them to know the profiles of players and referees have a better idea as to whether a racial biasness exists or not. Wolfers and Price do not have this type of access.

In my own conclusion, I have found that the Wolfers/Price study has done nothing more than to stir up a debate that is nothing more than water cooler talk at best.

Without any true other variables as far as other data collectors like interviews, eye witness accounts or even referee evaluations to corroborate their study, Wolfers and Price’s failure to examine the objectivity of the games called even in a time frame in which technology would be beneficial to them leads to suspicion of a bias result.

And while the same could be said for this article or even the Segal study, what is missing from the Wolfers/Price study are experts who have witnessed the games in question. It is easy to say that race plays a hand in so many aspects of our lives.

Yet the scientific community should be above reproach in questioning whether a company or private entity like the NBA is biased when there is evidence to the contrary.

As I stated earlier, in 15 seasons of covering the league, I have not seen any such actions where a referee had a racial bias towards any player I’ve covered. Even at just the coverage of home games, that’s hundreds of more games I’ve seen in person than any of these researchers have statistical information about.

Racial dynamic studies are important tools for not only science but also the corporate world as well. However, witch hunts, faux research papers and/or unfounded statistics should not be a part of a world that is still quite confused on a still very touchy subject.

While the Wolfers/Price study may be brilliant work and indeed it may be a paper that serves the corporate work place. But the paper paints with a very broad, uncontrollable brush on a private entity that over the past 12 decades has had a stellar record of improving racial harmony amongst management and the workplace.

This study is off base and that’s not the presumption from a fan of the NBA; that’s the professional opinion of a sports writer who has covered the league for 15 years and has personally seen more games either in person or on television than the researchers have spent hours putting their paper together.

EDITOR’S NOTES: Dr. Lapchick’s study can be found at http://www.bus.ucf.edu/sport/public/downloads/2005_racial_gender_ report_card_NBA.pdf.

The Wolfers/Price study can be found at http://graphics.nytimes.com/packages/pdf/sports/20070501-wolfers-NBA- race-study.pdf.