This study examined the association and predictive ability of several markers of internal workload on risk of injury in high-performance junior tennis players. Fifteen young, high-level tennis players (9 males, 6 females; age: 17.2 ± 1.1 years; height: 178.5 ± 8.7 cm; mass: 68.1 ± 4.8 kg) participated in this investigation. Data on injury epidemiology and internal workload during training were obtained for one competitive season. The session-rating of perceived exertion (s-RPE) was used to calculate internal workload markers in absolute (acute workload and chronic workload for 2-weeks, 3-weeks and 4-weeks) and relative terms (acute:chronic workload ratios [ACWR] for 2-weeks, 3-weeks and 4-weeks). Associations and diagnostic power for predicting tennis injuries were examined through generalised estimating equations and receiver operating characteristics analyses. During the season, a total of 40 injuries were recorded, corresponding to 3.5 injuries per 1000 h of tennis practice. The acute workload was highly associated with injury incidence (P=0.04), as injury risk increased by 1.62 times (95% CI: 1.01–2.62) for every increase of 1858.7 arbitrary units (AU) of the workload during the most recent training week. However, acute workload was a poor predictor of injury, and associations between injury and internal workload markers were weak (all P>0.05). These findings demonstrate an association between high values of acute workload and the risk of injury in high-level tennis players. However, a high acute workload is only one of the many factors associated with injury, and by itself, has low predictive ability for injury.