ABSTRACT
The purpose of this study was to build and test a demand model from game-related attributes to estimate fluctuations in attendance on a game-specific basis by controlling for the home team and examining those attributes solely associated with the visiting teams. Four major factors were selected to be the main focus of the current model, namely game schedule, star power, team performance, and uncertainty of outcome. The proposed model was tested using an empirical data set of a total of 1,189 NBA games played by six teams over five seasons. The resulted prediction models may be useful for estimating the demand of attendance because this study confirmed the strong effect of star power on demand, clarified the three-level day-of-week variable, added and verified the effect of the game availability variable, incorporated and validated new and more accurate variables to capture team performance, and discovered the potential moderating effect of the home team’s quality on the relationship between uncertainty of outcome and game demand.
