The impact of weakness and recovery time on recurrent injuries in NBA: a dynamic event-count model with transient carryover effect

Abstract The recurrence of injuries represents a major challenge in professional sports, with relevant financial implications for teams and considerable psychological impact on athletes, highlighting the importance of adopting sustainable risk management strategies. This paper aims to evaluate the main factors affecting repeated injuries among National Basketball Association (NBA) players within a recurrent-event analysis framework based on a dynamic event-count model. For this purpose, the study relies on three big datasets regarding players and games over a ten-year period, each characterized by complex structures. A key contribution of this work is the introduction of the Weakness , a variable designed to model the momentary vulnerability of players due to previous injuries and to capture a carryover effect of past injuries on future risk. This measure reflects the increased vulnerability of a player when multiple injuries occur within a short period, with the effect gradually decreasing as the time without new injuries increases. Beyond Weakness , both recovery time and player-specific attributes were investigated as potential factors determining the risk of relapse. Empirical results show that Weakness and recovery time significantly affect the risk of injury recurrence. Furthermore, age and playing position emerge as additional risk factors. The evidence highlights the potential of data-driven models to inform decision making in sports management, supporting sustainable strategies to reduce injury risk and enhance athlete availability, ultimately contributing to the long-term sustainability of their performances.