This one-year longitudinal study examined trait emotional intelligence as a predictor of Internet gaming disorder (IGD). To date, only cross-sectional research has been conducted to test the protective effects of emotional intelligence against IGD tendency. Based on the Interaction of Person-Affect-Cognition-Execution (I-PACE) model, this study aimed to address the research gap by examining not only the direct effects of trait emotional intelligence, but also its indirect effects (via depressive symptoms and coping flexibility) on IGD, with both a cross-sectional and longitudinal design. The participants were 282 Chinese university students (mean age = 20.47; 39.4% males) who voluntarily completed an anonymous questionnaire at both baseline (W1) and one-year follow-up (W2). Path analysis results revealed that trait emotional intelligence had a protective but indirect effect on IGD tendency in both our cross-sectional and longitudinal data. Depression was found to have a significant, full mediating effect on the relationship between: (i) trait emotional intelligence and IGD tendency (W2) and (ii) coping flexibility and IGD tendency (W2), after adjusting for IGD tendency at the baseline (W1). Gender invariance of the path coefficient was also observed in the prospective model. This study provided longitudinal evidence to support the I-PACE model. Interventions should address both IGD and depressive symptoms, and school-based workshops to increase emotional intelligence and coping flexibility are also recommended.