Segmenting Major League Baseball Teams by Attendance: A Multilevel Analysis of Determinants Across Clusters

Abstract

The study examines determinants of game-day attendance in Major League Baseball (MLB) by classifying teams based on their seasonal average attendance and analyzing how established predictors operate within each attendance segment. Using data from all 30 MLB teams across the 2006–2017 regular seasons (N=29,056), the analysis proceeded in two stages. First, using cluster analysis, the study identified four attendance-based groups that reflected distinct market and performance profiles. Second, three-level hierarchical linear modeling (HLM) evaluated the effects of economic, demographic, game-attractiveness, and residual-preference factors on single-game attendance within each group. Results revealed meaningful differences across clusters. High-attendance teams were less sensitive to opponent characteristics and more influenced by structural factors such as stadium capacity and home team performance. Lower-attendance teams showed stronger responses to visiting team quality, star players, game uncertainty, and local market conditions. Group-specific patterns also emerged for rivalry games, weekend scheduling, and seasonal progression. Findings demonstrate the value of segmenting MLB teams according to historical attendance patterns and highlight the utility of multilevel modeling for analyzing sport consumption. Results provide practical implications for MLB organizations seeking to tailor marketing strategies, improve scheduling decisions, and enhance fan engagement across diverse market environments.