Competing against the same team: does the length of time between the games matter? Evidence from the UEFA Champions League

There are many examples in professional sports in which two teams compete several times against each other over a short period of time. This structure raises many questions, including: Does the winner of the first match have a higher win probability in the second match? Does the length of time between these matches have an effect on win probabilities? Does it matter whether the winner of the first match was the underdog or the favorite? These questions are mostly asked by sports fans, journalists and coaches. However, repeated games may be of much broader interest among economists, psychologists and others. This is because repeated interactions occur very often in many areas of life, such as politics, the labor market, tenders, etc. Therefore, insights obtained from sports competitions may also be useful in other domains. In this study, I take advantage of real-life contests between professionals with large monetary rewards, where each pair of contestants competes against each other twice. However, some pairs compete twice in a row in two consecutive matches, whereas other pairs compete against each other for the second time after a longer period of time. Moreover, the allocation of pairs into treatment groups (playing the return match after a short period of time) and control groups (playing the return match after a long period of time) was determined via explicit randomization, which simplifies credible causal inference (Manski, 1995).