Heterogeneity in Mixed Strategy Equilibrium : Evidence from the Soccer Field

I conduct a (quasi) field experiment to study if individuals satisfy the implications of mixed strategy equilibrium when they repeatedly face the same opponents. An important advantage of my dataset is that it contains repeated observations on heterogeneous pairs of soccer players, a situation that rarely repeats in real life. In this respect, I show that assuming homogeneity of opponents leads to false rejections when opponents are in fact heterogeneous, and suggest valid aggregate tests that combine statistics across opponents. Empirically, I find that the behavior of most players is consistent with the theory. However, when I treat heterogeneous rivals as homogeneous, I find statistical differences in the scoring probabilities across strategies as well as serially correlated actions