This paper provides empirical evidence of favoritism by agents, where that favoritism is generated by social pressure. To do so, we explore the behavior of professional soccer referees. Referees have discretion over the addition of extra time at the end of a soccer game (called injury time), to compensate for lost time due to unusual stoppages. We test for systematic bias shown by Spanish referees in favor of home teams. We show that referees systematically favor home teams by shortening close games where the home team is ahead, and lengthening close games where the home team is behind. They show no such bias for games that are not close. We further show that when the rewards for winning games increase, referees change their bias accordingly. We also identify that the mechanism through which bias operates is the referees' desire to satisfy the crowd, by documenting how the size and composition of the crowd affect referee favoritism.