Experience shapes non-linearities between team behavioral interdependence, team collaboration, and performance in massively multiplayer online games.

This paper examines quantitative predictors of team performance in Massively Multiplayer Online Games (MMOGs) based on team management literature. Analyzing data from more than 140,000 squad-mode matches involving over 500,000 players, we replicate and extend existing research by confirming a curvilinear association between behavioral interdependence and team performance and introduce the moderating effect of experience. For less experienced teams, behavioral interdependence follows an inverted U-shaped pattern showing that excessive collaboration may be counterproductive. However, this is not the case for experienced teams, where the relationship is fairly linear. Additionally, we observe that riskier teams tend to perform worse. Moreover, our research also highlights the potential of e-sports data in advancing behavioral science and management research. The digital nature of e-sports datasets, characterized by size and granularity, mitigates concerns related to reproducibility, replicability, and generalizability in social science research, offering a cost-effective platform for scholars with diverse backgrounds.