Effects of different velocity loss thresholds during resistance training on jump and sprint performance in trained populations: a systematic review and three-level meta-analysis

Purpose To systematically evaluate the effects of different velocity loss (VL) thresholds during resistance training on jump and sprint performance in trained populations and to examine potential moderators. Methods PubMed, Web of Science, Embase, the Cochrane Library, and SPORTDiscus were searched from inception through April 10, 2026. Randomized controlled trials meeting the prespecified PICOS criteria were included. A three-level random-effects meta-analysis was conducted to calculate Hedges’ g and 95% confidence intervals (CIs), accounting for dependence among multiple effect sizes within studies. Risk of bias and certainty of evidence were assessed using RoB 2 and GRADE, respectively. Results Twelve randomized controlled trials comprising 42 effect sizes were included. Lower VL thresholds produced a small but statistically significant favorable effect on combined jump and sprint performance compared with higher VL thresholds (g=0.29, 95% CI: 0.17 to 0.42, p<0.001), with negligible heterogeneity ( I total 2 = 0.1 % ). Domain-specific analyses showed significant favorable effects for both jump performance (g=0.27, 95% CI: 0.10 to 0.45, p=0.002) and sprint performance (g=0.32, 95% CI: 0.13 to 0.50, p<0.001). Outcome-specific analyses indicated clearer effects for countermovement jump, 10 m sprint, and 30 m sprint, whereas the estimates for squat jump and 20 m sprint were not statistically significant. Significant effects were observed for the 10% versus 30% VL and 10% versus 20% VL comparisons, whereas the 20% versus 40% VL comparison yielded a smaller, non-significant effect. Exploratory moderator analyses did not detect statistically significant effects of population type, training frequency, training mode, intervention duration, or VL-threshold difference; however, these analyses may have been underpowered because of the limited number of included studies. The direction and statistical significance of the pooled effect were robust across alternative pre–post correlation assumptions and model specifications, although heterogeneity estimates and prediction intervals were more sensitive to higher assumed correlations. No evidence of significant small-study effects was detected. Conclusions Lower VL thresholds may confer small favorable effects on jump and sprint performance in trained populations, with clearer evidence for countermovement jump and selected sprint outcomes. However, the effects were not consistent across all outcomes, and the certainty of evidence ranged from low to moderate. Lower VL may therefore be considered as one potential option for speed–power-oriented resistance training rather than a universally superior prescription. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/ , identifier CRD420261396083.