Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Previous research exploring the role of belief dynamics for consumers in the entertainment industry has largely ignored the fact that emotional reactions are a function of the content and a consumer’s disposition towards certain protagonists. By analyzing 19 m tweets in combination with in-play information for 380 football matches played in the English Premier League we contribute to the literature in three ways. First, we present a setting for testing how belief dynamics drive behavior which is characterized by several desirable features for empirical research. Second, we present an approach for detecting fans and haters of a club as well as neutrals via sentiment revealed in Tweets. Third, by looking at behavioral responses to the temporal resolution of uncertainty during a game, we offer a fine-grained empirical test for the popular uncertainty-of-outcome hypothesis in sports.