Today’s guest blogger is Rob Simmons. Rob is a Senior Lecturer in Economics (in the U.S. this would be an associate professor) at Lancaster University, UK. One of the more prolific sports economists in Europe (and the world), Rob is on the editorial board of both the International Journal of Sports Finance and the Journal of Sports Economics. Rob is also presently working with Dave Berri on pay and performance relationships in NFL. Well, Rob is working. Dave is busy avoiding his work, hence the need for Rob to write a post today.
Results from a new study — co-authored by Babatunde Buraimo (University of Central Lancashire), David Forrest (University of Salford), and myself — will be presented on Saturday, May 12 at a conference on sports economics to be hosted by the Swiss Sports Institute in Magglingen.
Academic work on referee bias in sports has received considerable attention in the last few years. For example a new American paper, by Justin Wolfers and Joseph Price on racial bias in calling fouls in basketball, has been vigorously debated after details were recently revealed in the New York Times.
In European soccer, work has concentrated on home team bias by officials.
Sir Alex Ferguson said, about the referee’s performance in Manchester United’s away match in Rome last month, that “it was like playing against twelve men”. But is it generally true that referees tend systematically to favour the home team? Earlier published work says ‘yes’. For example, analysis of the length of stoppage time added to matches in Premier Leagues in Spain, Germany, Italy, the US and England yields evidence that much more time is added when the home team is a goal behind rather than a goal ahead.
Laboratory work also supports the hypothesis. An earlier study asked English referees to adjudicate on whether tackles shown on video were fouls. A group shown film with crowd noise consistently called more decisions in favour of the home team compared with other referees who viewed without a soundtrack. Referees appeared to be influenced by the crowd.
Our new paper looks for evidence of home team bias in the handing out of yellow and red cards. Home teams receive fewer cards. But this might not result from referee bias. Away teams are often the underdogs and their game plan may be to defend vigorously, producing more fouls. Similarly, away teams trail in a game more often than home teams….and teams that are behind may get desperate and commit more offences, again probably incurring punishment from the referee.
Any attempt to identify referee bias has therefore to allow for the relative strengths of the two teams and what goes on in the game.
To examine this issue we obtained information from over 200,000 individual minutes of play in the English Premier League and nearly 160,000 minutes in Bundesliga 1. We then built a statistical model to account for the factors that generate home and away yellows and reds in any given minute of a football game. This quantifies, for example, how much more likely cards are in a derby match and it illustrates the impact on probabilities of cards in the remainder of a game when one team takes the lead or indeed receives a yellow or red.
Many of these findings will interest soccer fans; but the purpose of such detailed analysis is to find out whether, even after allowing for a rich variety of factors relevant to the handing out of red and yellow cards, there is still a tendency for home teams to be punished less often.
The answer is “yes”:
In both England and Germany, home team players are less likely to receive a yellow card.
In both England and Germany, there appears to be much more reluctance to send off a home player than an away player.
Intriguingly, for Germany, lenient treatment of home players is much less pronounced when the sample includes only matches at stadia where the field has a running track around it. This is suggestive that crowd influence on referees is reduced if the crowd is well back from the action. So our analysis suggests that referees can indeed be intimidated. And the level of intimidation is influenced by of all things, how the stadium is constructed.
For those who want more details, this paper will be presented at the Western Economic Association in Seattle in July.
– Rob Simmons