Background and Objectives: This study aimed to assess the week-by-week predictive value of Acute:Chronic Workload Ratios (ACWRs) for non-contact injury risk in professional soccer players. Materials and Methods: A cohort of 40 elite players was monitored using GPS over two competitive seasons. Binomial logistic regression and ROC curve analyses were performed on ACWR metrics—including total distance, moderate-to high-speed running, sprinting, acceleration, and deceleration—during the four weeks prior to injury (W4 to W1). p-values were further adjusted for multiple comparisons using the false discovery rate (FDR) correction (q < 0.05). Results: Significant predictive models emerged mainly for ACWR metrics related to moderate-speed running (15–20 km/h), sprinting (>25 km/h), and acceleration/deceleration. The ACWR for 15–20 km/h (DSR15–20) demonstrated the highest predictive accuracy, particularly in Week 3 (AUC = 0.811, p = 0.004). Sprinting (DSR>25) was also significantly associated with injury occurrence across Weeks 1–4 (AUC = 0.709–0.755, p = 0.011–0.024). Acceleration (ACC) and deceleration (DEC) metrics showed significant associations prior to correction—ACC in Weeks 3–4 (AUC = 0.737–0.755, p = 0.020–0.026) and DEC in Weeks 3–4 (AUC = 0.720–0.741, p = 0.029–0.043)—but these associations did not retain significance following FDR adjustment (q = 0.052–0.086). In contrast, total distance (ACWR TD) and high-speed running (20–25 km/h) were weaker predictors, reaching only marginal or nonsignificant levels (e.g., Week 3, AUC = 0.675, p = 0.054). After FDR correction, only DSR15–20 and DSR>25 remained statistically significant (q < 0.05), confirming them as robust predictors of non-contact injury risk. Multivariable models adjusted for age and playing position confirmed these findings, with DSR15–20 and DSR>25 retaining their predictive value independent of confounding factors. Injury risk thresholds were established through Estimated Marginal Means (EMMs), defining the “Sweet Spot” and “Danger Zone” for each metric, whereas the “Low Load” zone was treated as exploratory. Conclusions: This weekly ACWR monitoring approach enables practical injury risk profiling, helping training staff optimize load management and minimize non-contact injury risk in elite soccer.

This study examined the effects of two different intensities of running between repeated sprints and compared them with passive recovery. Thirteen professional soccer players performed two sets of six 30 m sprints on three randomly assigned occasions. A 5 min passive rest period separated the two sets, while sprints were interspersed with either passive standing, running at 95% of the first lactate threshold (MOD) and running at maximum aerobic speed (HIGH). Performance decrements were greater in HIGH than MOD at the last sprint in both sets (set 1: 5.8 ± 4.2% vs. 2.6 ± 3.2%, p = 0.07; set 2: 9.1 ± 4.5% vs. 4.0 ± 6.1%, p = 0.016). Acceleration (0–15 m) was more affected than maximal-speed running (15–30 m) (condition × sprint interaction: p < 0.001). Mean and peak heart rate were higher in both running conditions than passive (p < 0.05), with no difference between MOD and HIGH. Blood lactate showed a significant set × condition interaction (p < 0.001), peaking at 13.6 ± 2.7 mmol·L⁻¹ in HIGH, while blood lactate responses to passive and MOD were similar and peaked after the second set of sprints (10.7 ± 2.1 and 11.5 ± 2.8 mmol·L⁻¹, respectively). Between-sprint running intensity markedly influenced fatigue development during repeated-sprint exercise. The HIGH condition elicited greater metabolic strain and performance decrements than MOD or passive conditions. Within the present protocol, passive standing was associated with smaller decrements in repeated-sprint performance despite high heart rate and blood lactate responses.