Uncertainty of Outcome Hypothesis: Theoretical Development and Empirical Evaluation

This paper adopts both a theoretical and an empirical approach to the topic of outcome uncertainty (OU) in sports. First, it develops a standard quality-based demand model in which a fan has utilities from match quality and team quality. It theoretically demonstrates that fan attendance does not necessarily exhibit an inverted U shape, even in the presence of OU. It suggests two different attendance regression model specifications: the “standard” model, which estimates the effect of OU separately by controlling for the effect of team quality, and the “alternative” model, which estimates the match and team quality effects together. Second, the study evaluates earlier empirical studies and finds that all applied the standard model, and they often failed to completely control for the team quality effect. Third, it conducts empirical exercises with two daily attendance samples from the English Premier League (EPL) and Major League Baseball (MLB) and two annual attendance samples from the Korean Baseball Organization and MLB. It demonstrates that the standard model regression results are also sensitive to model specification. However, the alternative model is exempt from the identification issue, and its regression results are consistent across different specifications.