Skill, Strategy and Passion: an Empirical Analysis of Soccer

Sports provide a natural experiment on individual choices in games with high stakes. We study soccer, analysing a simple game theoretic model of a match and using data to evaluate the performance of this model in explaining actual behaviour. A team's optimal strategy depends on the current state of the game. When the game is tied, both teams attack because this maximises their probability of scoring. A losing team always attacks, while its winning opponent may attack early in the game, but starts defending as the end of the match nears. We estimate the probability of scoring in a soccer match using data from the Italian, English, and Spanish leagues. Teams' skills and playing at home are signi ̄cant explanatory variables. We also ̄nd this probability to be a®ected by the current score of the match. Finally, losing teams are more likely to score, and winning teams adopt conservative strategies very early in the match. Although these ̄ndings support the main conclusions of our model, and indicate soccer teams' behaviour is consistent with rationality, we also ̄nd evidence emotional factors a®ect their decision making.