Academic Performance Predicts Competitive Gaming Talent: A Time-Adjusted OLS Analysis of CS2 Players

Abstract

This study examines the relationship between academic performance and gamine talent in competitive video games by focusing on competitive game Counter Strike 2 (CS2) and separates practice-time effects from individual ability. Data were collected from 51 male senior-2 students in Haidian District, Beijing, who met minimum equipment and engagement thresholds. Two ordinary least squares (OLS) approaches were applied: (1) an adjusted-dependent-variable method that divides raw performance by a logarithm of game duration term to approximate performance efficiency, and (2) a residual method that first regresses raw performance on logarithm of game duration and then regresses the resulting residuals (the time-adjusted “talent” measure) on academic score. Results show a consistent and significant positive relation between final exam scores and time-adjusted gaming performance. In the adjusted-Y model, the final score coefficient is positive (coefficient = 1.15) and significant (p < 0.01), with the model explaining about 20% variance. In the residual approach the relationship remains significant (p < 0.01), and the second regression of the model explains roughly 26% of variance. These findings indicate that, beyond practice time, higher academic achievement is associated with greater efficiency or “talent” in competitive gaming. The research findings challenge the simple stereotype linking gaming proficiency to poor academic performance and suggest that cognitive and learning abilities may have cross disciplinary transfers.