Maximizing player value through the application of cross-gaming predictive models

Purpose – This paper aims to, considering the potential to generate additional revenue from cross-gamers, identify variables predicting predominant slot-players’ propensity to play table games, as well as predominant table-game players’ propensity to play slots (cross-game play). Casino marketers often promote cross-game play through game lessons and coupons for game trial. Design/methodology/approach – Logistic regression analysis was performed on the player data provided by a destination hotel casino on the Las Vegas Strip. Furthermore, the authors described how to estimate propensity scores, the probability of cross-game play, at the individual level, using a logistic regression equation. Findings – Comparisons of cross-gamers versus non cross-gamers indicated that the amount of play and gaming values of cross-gamers were much higher than those of slot-only players. The results of a logistic regression analysis show that a player’s cross-gaming propensity can be predicted using gaming-related behavioral data. More specifically, cross-gaming propensities were associated with the frequency and recency of casino trips, the amount of money won or lost in gaming, player values to the casino, the duration of play and the length of a customer–casino relationship. Research limitations/implications – It is recommended that future research apply the model tested herein to other samples and investigate other predictor variables to develop a better predictive model for cross-game play. Practical implications – The findings and the model introduced herein could help casino marketers identify players with cross-gaming propensity and develop more targeted strategies for customer-relationship management and database marketing. Originality/value – This study is the first attempt to estimate the cross-gaming propensity at the individual level and offers detailed guidance on how to use the propensity scores for targeting specific customers.