TCS Daily


Refereekonomics

By Raymond Sauer - August 2, 2006 12:00 AM

Referee decisions always seems to be in the news. Disputes over the calls made by umpires, linesmen, or refs in the game of your choice have surely been part of the conversation since the beginning. But just as surely, the advantages given television viewers and commentators with the advent of instant replay have escalated the rhetoric. In 2006 alone, allegations of egregious officiating errors have been made regarding the Super Bowl, the NBA Finals, and the World Cup -- that is, in about every high-profile championship you can imagine.

Officiating errors will never be eradicated -- in that sense they, like the mistakes that athletes make, are part of the game. But it surely makes sense to take cost-effective actions to reduce the impact of referee mistakes on the outcomes of sport. To the greatest extent possible, fans want results to be determined by the athletes on the field of play, and not some knucklehead in black and white stripes.

Yeah, alright you say. But what does this have to do with economics? As a card-carrying member of the economic imperialist club, I believe economics has something to say on the topic; and the minions that manage sports would benefit from listening. In fact, there are several areas of impact. In this essay we'll focus on two: referee bias, and the effect of incentives on the decisions of referees and players.

That referees are biased would appear to be a mug's claim, but it's true. It has long been argued, for example, that referees tilt their decisions in favor of the home team. Many statistics point in this direction, like the foul differential between the home and visiting team in a basketball game ( but these don't always offer a clean interpretation). Recently, however, several groups of economists have identified conditions which lend to a clean test: the length of time added on ("injury time") at the end of a football (soccer) match.

What these researchers find, using data from soccer leagues in Spain, Italy, Germany, and the U.S., is that referees add more time to the game when it matters for the home team, particularly when the home team is trailing in a close contest. When the home team is winning a close game or the game is out of reach, the refs blow their whistles sooner. In the case of La Liga in Spain, the extra additional time is almost two minutes in a close contest! The authors interpret this as convincing evidence of referee favoritism towards fans of the home team. And I agree.

Of particular interest is a paper which documents changes in the home bias of football referees, by focusing on a recent change in incentives. Neil Rickman and Robert Witt of the University of Surrey measure the time (pdf) of English matches surrounding the 2000-2001 season. During this season, England's top division in football -- the Premier League -- adopted a policy of hiring professional referees. By paying the best referees a premium and monitoring their decisions closely, referees have a financial stake in calling the game "straight." The remaining English divisions continued their age-old practice of hiring amateur refs.

Rickman and Witt find that home bias in England is less in general than that in Spain. But in particular, injury time during close games in the Premier League decreased by half a minute relative to prior periods and other leagues. The lesson is that referees respond to incentives, and active monitoring of referee decisions by the league has an impact.

Players respond to incentives as well. Controversy flared during the World Cup over players simulating fouls to gain strategic advantage by players. This is not a new issue, but there may be a straightforward solution. Prior to the World Cup, my colleague Brian Goff argued that FIFA should allow post-match video review to punish such offenses after the game is played. The logic in Goff's proposal stems from one of the original papers in Sports Economics, "Crime on the Court," by Robert McCormick and Robert Tollison.

McCormick and Tollison studied an experiment by the Atlantic Coast Conference (ACC) in the 1978 season, when they added a third referee to call college basketball games. The added referee implies that a greater percentage of the fouls that take place will be detected, but the number of fouls called actually decreased. Why? Because with the third ref, players knew that their transgressions were more likely to be observed. Increasing the probability of detection (given the appropriate punishment) will decrease the amount of undesired behavior in a sport, as is surely the case with simulated fouls in soccer. The third referee is now a regular feature of NCAA and NBA basketball precisely for this reason. When it comes to referees, incentives matter.

Skip Sauer writes The Sports Economist column for TCS Daily. Find more of his writing here at The SportsEconomist.com.

Categories:

4 Comments

You are unfair
As an official in the Southwest Football Officials Association, I never saw bias for the home team. Most people have little knowledge of how hard officials plan before games to ensure fairness by players and prevent injuries, along with weekly meetings to discuss rule controversies. Officials take the game serious and want to call a game with no errors, if possible. But, officials are humans and are not perfect. We wanted to be, but it is not possible. Officials are trained, however, to not have rabbit ears and listen to the crowd. In most cases these persons do not know the rules nor are they on the field to witness the fouls. Even the TV camera only focuses on the ball and never the individual players holding, slugging or tripping. Most fouls occur away from the ball.

I think your article is unfair and crowd biased. Of course, what is new? We were use to it. I quit a hobby I loved not because of accusations such as yours, but because my job just simply would not allow it. I can remember times when crowd booing was going on and the players on the field were in fact embarrassed, especially when the Coach was making a you know what out of himself. Again, I loved it, and it was not for the money. It is a shame you think all officials cheat for the home team. It is not true.

crowd biased?
It is fine to call me unfair, but what is your explanation for the two minutes of added time awarded to home teams when they trail by a goal?

The explanation offered by economistsm who have measured this, is that the ref is influenced by the crowd. An effective monitoring and reward system has been shown to address this problem.

This is an economic point. I did not intend the piece to be another "lets crack on the stupid ref" tirade.

Problem is you lump all sports into one (the basketball example does not count)
I still think the economistsm (your spelling) is only looking for what they want to see. In addition, the idea that statistics validates their conclusion can be biased as well. Read "Damn Lies and Statistics" by Joel Best. I also have used statistics in engineering. What is it you want to prove? Unless I can see the sample used for the statistical analysis, I am always skeptical. I also know sports officials too and have worked in this area. Have you?

my apologies for the mis-spelling
The stats on added time have been published in peer-reviewed journals and replicated by subsequent teams of researchers. You are free to choose, so ignore these results if you wish.

For the record, I'm a former youth coach who detests ref-baiting at all levels of sport. But personal experience is not the point here. Whether I have worked the other side of the game does not change the facts that I reported in the piece.

TCS Daily Archives