The number of studies on the relationship between training and competition load and injury has increased exponentially in recent years, and it is also widely studied by researchers in the field of professional soccer. In order to provide practical guidance for workload management and injury prevention in professional athletes, this study provides a review of the literature on the effect of load on injury risk, injury prediction, and interpretation mechanisms. The results of the research show that: (1) It appears that short-term fixture congestion may increase the match injury incidence, while long-term fixture congestion may have no effect on both the overall injury incidence and the match injury incidence. (2) It is impossible to determine conclusively whether any global positioning system (GPS)-derived metrics (total distance, high-speed running distance, and acceleration) are associated with an increased risk of injury. (3) The acute:chronic workload ratio (ACWR) of the session rating of perceived exertion (s-RPE) may be significantly associated with the risk of non-contact injuries, but an ACWR threshold with a minimum risk of injury could not be obtained. (4) Based on the workload and fatigue recovery factors, artificial intelligence technology may possess good predictive power regarding injury risk.

This study aimed to investigate whether the heart rate variability index (TLHRV) during five ball-drills could be used to quantify training load (TL) in collegiate basketball players. Ten elite male college basketball athletes (18.2 ± 0.4 years) were recruited to perform five ball-drills (1V1, 2V2, 3V3, 4V4, and 5V5) which lasted 10 min and varied in intensity. During each drill, TLHRV, training impulse (TRIMP), rating of perceived exertion (RPE), speed, and distance were recorded by Firstbeat, Foster’s RPE scale, and SiMi Scout. The correlation (Spearman’s and Pearson’s correlation coefficient), reliability (intra-class correlation coefficient, ICC), and agreement (Bland-Altman plots) among TLHRV, TRIMP, RPE, speed, and distance were examined. TLHRV was significantly correlated with TRIMP (r = 0.34, p = 0.015) and RPE (r = 0.42, p = 0.002). TLHRV was significantly correlated with training intensity (r = 0.477, p = 0.006) but not with volume (r = 0.272, p = 0.056). TLHRV and TRIMP, RPE showed significant intraclass relationships (ICC = 0.592, p = 0.0003). Moreover, TLHRV differentiated basketball drills of equal volume and varying intensity. We concluded that TLHRVmay serve as an objective and rational measure to monitor TL in basketball players.