Although role stress has been much investigated with regard to performance in organizational contexts, meta-analytic support for a robust relationship between role conflict and performance is surprisingly scarce. The present study considers how the experience of role stress (i.e., the presence of unclear and/or conflicting role expectations) may influence performance in sport teams characterized by a high degree of task interdependence. Based on Kahn, Wolfe, Quinn, and Snoek’s (1964) role episode model, we hypothesized that conflicting information regarding one’s role would detract from one’s performance through its positive association with role ambiguity. Club and university interdependent sports team athletes (N = 243) from 30 teams and their respective head coaches participated in the study. Athletes identified their four primary roles and subsequently completed a questionnaire assessing perceptions of role conflict and the scope of responsibilities subscale from the Role Ambiguity Scale (Beauchamp, Bray, Eys, & Carron, 2002). To reduce some of the concerns associated with common method variance, an informant rating of role performance was obtained via each athlete’s head coach. Multi-level analysis (MLwiN; Browne, Rasbash, & Charlton, 2012) was employed to model the nested nature of the data. Results revealed that the predicted indirect effect was significant. More specifically, we found a positive association between role conflict and role ambiguity. Conversely, role ambiguity was negatively related with role performance. The findings offer support for the influence of role conflict—albeit through an indirect manner—and suggest that it would be remise of applied practitioners to ignore its impact within sports teams. Future research should continue to investigate the experience of role stress and consider if additional mediating variables (e.g., collective efficacy) are relevant to the role conflict—role performance association.