Abstract
This paper develops a new simulation-based measure of playoff uncertainty and investigates its contribution to modelling match attendance compared to other variants of playoff uncertainty in the existing literature. A model of match attendance that incorporates match uncertainty, playoff uncertainty, past home-team performance and other relevant control variables is fitted to Australian National Rugby League data for seasons 2004-2008 using fixed effects estimation. The results suggest that playoff uncertainty and home-team success are more important determinants of match attendance than match uncertainty. Alternative measures of playoff uncertainty based on points behind the leader, although more ad hoc, also appear able to capture the effects of playoff uncertainty.