Notes: Estimates as indicated by column; average marginal effects; (d) for discrete change of dummy variable from 0 to 1. The dependent variable is a dummy variable that takes value 1 for wrong calls among penalties and goals that should be awarded. Debatable decisions are excluded. The regressions include controls for the home team, win probability (odds based), referee experience and average performance in the season excluding the current game, the last 10 min of the game, closeness of the game, decisiveness of the call, the number of spectators, accuracy of previous decisions, and stadiums with running tracks. Furthermore, season, match day, referee, and team-fixed effects are included. Standard errors in parentheses are clustered at the game level except for the Firthlogit regression, which does not allow for clustering. Note that the logit regression struggles at estimating some referee-fixed effects and hence excluded two referees from the sample, explaining the slightly smaller observation number.

p < 0.01,

p < 0.05,

p < 0.1.