OBJECTIVES: Our primary aim was to determine if athletic identity is prospectively associated with shoulder overuse injuries. Secondly, we aimed to determine if athletic identity is prospectively associated with playing through pain and to describe how athletic identity relates to sex, age, playing level, weekly training load, and match volume. METHODS: A cohort of 269 adolescent tennis players were followed over a period of 52 weeks. Cox regression was used to estimate the hazard rate ratio (HRR) of first-time shoulder overuse injury associated with every 10-unit increase on the Athletic Identity Measurement Scale (AIMS). RESULTS: The adjusted HRR of shoulder overuse injury was 0.89 (95% CI: 0.36-2.20) and the odds ratio of playing through pain was 2.41 (95% CI: 0.74-8.96) for every 10 unit increase on AIMS. The level of athletic identity was higher among players at the national level than among players at the regional level and was weakly correlated to weekly hours of tennis matches, tennis training, and fitness training. CONCLUSIONS: Our results indicate that higher levels of athletic identity may be associated with a lower incidence of shoulder overuse injuries, and potentially with playing through pain, although these results are inconclusive due to wide confidence intervals.