Referees are inherently biased
Sports referees are biased towards the home team, but not deliberately — the book review of “Scorecasting” explains. From The Economist, March 10th, 2011:
A new book looks at the behavioural economics of sport
“YOU want a fair referee, you know…You want a strong referee, anyway, and we didn’t get that,” grumbled Sir Alex Ferguson, the manager of Manchester United, an English soccer team, after a recent defeat at a rival’s ground. Sir Alex’s comments have landed him in hot water with the game’s governing bodies. In his defence, he would do well to cite “Scorecasting: The Hidden Influences Behind How Sports Are Played And Games Are Won”, a fascinating new book on behavioural economics and sport. Among other things it provides some compelling evidence of referee bias in favour of home teams.
The book is an attempt to test and explain a series of popular beliefs about sport by two statistics-obsessed fans, Tobias Moskowitz, an economist at the University of Chicago, and Jon Wertheim, a journalist with Sports Illustrated, a magazine better known for its swimsuit models than its economic models.
Sport is a multi-billion-dollar industry, they note, but its underlying economics are rarely studied. Yet it is in many ways the perfect laboratory for testing economic theories about decision-making. There is an abundance of readily available data, on everything from the consistency of a golfer’s putting to the frequency of use of certain types of tactics in American football. The goals of participants in sporting contests are relatively uncomplicated (score, win, enforce the rules) and the outcomes are extremely clear, both of which help with designing theoretical models. And the stakes are often high: the sums of money involved dwarf the small amount of dollars economists typically offer participants in their experiments.
The authors dismantle some sporting myths—the Chicago Cubs, a baseball team that has not won America’s World Series for over a century, are not “cursed”. But other bits of pub wisdom have substance. The authors find evidence of bias by officials when they try to answer the question why home teams win more often than away teams. There is nothing imaginary about this home advantage: the home team wins 54% of games in Major-league Baseball, 60% in international cricket, 63% in English Premier League soccer and 69% in American college basketball. The authors test various explanations, but find no evidence of crowd support inspiring home players to do better; no correlation with distance travelled to the game; little, if any, statistical support for the view that home teams win because of greater familiarity with the idiosyncrasies of their stadium (like the Green Monster, a wall at Fenway Park, the home of the Boston Red Sox).
What the evidence does show, however, is that referees favour home teams in judgment calls, particularly those that happen at a crucial stage in a game. The authors focus on decisions that are less susceptible to influence by player pressure, but give some discretion to the official. If a batter chooses not to swing at a baseball pitch, the pitch is more likely to be called a strike if the home team is pitching. This tendency is most extreme in close games. In soccer, referees are more likely to award penalties to the home team, hand out fewer punishments for offences to home players, and adjust the amount of overtime played depending on whether the home team is losing (more overtime is given) or winning (less). Sir Alex is unlikely to dwell on this last point: the extra minutes that Manchester United get at home when they are trailing is known as “Fergie time”.
Are referees deliberately biased? The authors think not. Instead, they blame the fact that referees, like the rest of us, tend subconsciously to rely on crowdsourcing, picking up on the mood of the crowd when making their decision. “Anchoring” is the name economists give to people’s tendency to be unduly influenced by outside suggestion. Take away the crowd and the home bias shrinks, as it did a few years back when 21 Italian soccer matches were played without supporters following incidents of crowd violence. In these games the home bias declined by 23% on fouls called, by 26% for yellow cards and by a remarkable 70% for red cards, which remove a player from the game and have a particularly big impact on the result. This bias is consistently higher in soccer than in most other sports because there are many more judgments for the referee to make. If technology is ever adopted to help referees—using TV to review contentious calls, for instance—away teams should benefit most.
Psychological biases influence decision-making by players as well as officials. “Loss aversion”, the tendency for people to care more about avoiding a loss than about making a similar-sized gain, is rife even among the greatest champions. Evidence from a huge database of near-identical putts shows that, along with all the other golfers studied, Tiger Woods is more likely to hole a putt if it is to save par (in golf, an over-par hole feels like a failure) than if he had the identical putt to make a birdie (a gain). The authors argue that this is because he tries harder to avoid the loss than he does to make the gain. That makes no sense—the score versus par on an individual hole is not what matters in golf, but the number of strokes taken over the course of 18 holes. Each putt counts exactly the same, yet players treat them differently.
Another common form of behaviour is managing complexity with simple rules by, for instance, giving disproportionate weight to round numbers. In baseball a batting average of .300 (reaching first base with a hit 300 out of 1,000 goes) is seen as the difference between a decent batter and an elite one. On the last day of the past 35 seasons a batsman with an average of .299 has never accepted a walk to first base (which does not help their average). Instead they swing at pitches and either get the hit that takes them to .300 or miss and get struck out, taking their average below .299. One extra successful hit in 1,000 attempts is a trivial difference. But it matters. The average salary of a batter hitting .300 exceeds that of someone hitting .299 by 2%, a juicy $130,000.
No comments yet.