<p>Cyberbullying and cybervictimization have become widespread phenomena on social media platforms, particularly in emotionally charged environments such as sports fandom. While prior research has emphasized the role of individual personality traits, the influence of social identification—specifically, fan identity—on these behaviors remains underexplored, especially within Chinese digital communities, such as Sina Weibo. Grounding the study in the Social Identity Model of Deindividuation Effects, we theorize that platform affordances, such as anonymity and group salience, amplify identity-congruent behavior in sports-fan interactions on Weibo. This study aimed to examine the predictive roles of Big Five personality traits (extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience) and fan identity in the occurrence of cyberbullying and cybervictimization among sports fans active on Sina Weibo. Our primary objective was to estimate the unique and incremental contribution of fan identity beyond personality variables and to clarify its relative importance in relation to bullying and victimization outcomes. A total of 500 Sina Weibo users who regularly participated in sports-related discussions completed an online questionnaire. Constructs were measured using validated scales (BFI-44, COAS, and SSIS-adapted). Partial Least Squares Structural Equation Modeling (PLS-SEM) was used to test the hypothesized relationships between personality traits, fan identity, and online aggressive behaviors. Data were collected in Mandarin; measurement adaptation procedures (use of validated Mandarin versions or translation/back-translation), item lists, and item-level psychometrics are reported in the Methods as well as the results. Model quality and predictive performance were evaluated using contemporary PLS-SEM criteria (e.g., HTMT for discriminant validity, VIF for collinearity, SRMR for model fit, and R²/Q² indices), and the choice of PLS over CB-SEM is justified on predictive and distributional grounds. Fan identity showed a robust positive association with both cyberbullying and cybervictimization, providing incremental explanatory power over and above the Big Five variables. The analysis revealed that extraversion was also significantly and positively associated with cyberbullying, while agreeableness had a significant negative effect. Neuroticism was the strongest predictor of cybervictimization, followed by a positive contribution from fan identity. Conscientiousness was negatively associated with victimization, while openness to experience showed no significant effect on either outcome. Both individual differences and group-level identification contribute to online aggression and vulnerability in sports-related discourse. By extending the SIDE framework to Chinese sports fandom on Weibo, the findings highlight identity salience as a practical lever for intervention (e.g., rivalry de-escalation norms and identity re-framing) alongside personality-informed supports; design limitations (cross-sectional, self-report, nonprobability sampling) are acknowledged, and future research should incorporate behavioral trace data and probability sampling.</p>