The aim of this study was to investigate the differences in neuromuscular performance between variable resistance training and constant resistance training within complex training. Twenty-one well-trained collegiate basketball players were randomly assigned to either an experimental group (variable resistance training) or a control group (constant resistance training) and completed a twice weekly training program over an 8-week period. Training programs were the same except that the experimental group included variable resistance via elastic bands (40% of the total load). Maximum strength, vertical jump, horizontal jump, and sprint performance were assessed pre- and post-intervention. Both groups demonstrated significant increases in the back squat 1RM (experimental group +36.5% and control group +32.3%, both p < 0.001), countermovement jump (experimental group +12.9%, p = 0.002 and control group +5.6%, p = 0.02), and squat jump performance (experimental group +21.4% and control group +12.9%, both p < 0.001), whereas standing broad jump performance improved only in the experimental group (+2.9%, p = 0.029). Additionally, the experimental group showed significant improvement in the squat jump (p = 0.014) compared with the control group. However, no statistically significant differences were found between groups for countermovement jump (p = 0.06) and sprint performance at 10 m (p = 0.153) and 20 m (p = 0.076). We may conclude that both training modalities showed similar improvements in maximum strength. Performing variable resistance training within a complex training program is more efficient to enhance selective power performance in well-trained collegiate basketball players.

Objective Acute effects of variable resistance training (VRT) and constant resistance training (CRT) on neuromuscular performance are still equivocal. We aimed to determine the differences between VRT and CRT in terms of force, velocity, and power outcomes. Methods We searched PubMed, Web of Science, and SPORTDiscus electronic databases for articles until June 2021. Crossover design studies comparing force, velocity, and power outcomes while performing VRT and CRT were included. Two reviewers independently applied the modified version of the Cochrane Collaboration’s tool to assess the risk of bias. A three-level random effects meta-analyses and meta-regressions were used to compute standardized mean differences (SMDs) and 95% confidence intervals. Results We included 16 studies with 207 participants in the quantitative synthesis. Based on the pooled results, VRT generated greater mean velocity (SMD = 0.675; moderate Grading of Recommendations Assessment, Development and Evaluation (GRADE) quality evidence) and mean power (SMD = 1.022; low) than CRT. Subgroup analyses revealed that VRT considerably increased the mean velocity (SMD = 0.903; moderate) and mean power (SMD = 1.456; moderate) in the equated loading scheme and the mean velocity (SMD = 0.712; low) in the CRT higher loading scheme. However, VRT marginally significantly reduced peak velocity (SMD = −0.481; low) in the VRT higher loading scheme. Based on the meta-regression analysis, it was found that mean power (p = 0.014–0.043) was positively moderated by the contribution of variable resistance and peak velocity (p = 0.018) and peak power (p = 0.001–0.004) and RFD (p = 0.003) were positively moderated by variable resistance equipment, favoring elastic bands. Conclusions VRT provides practitioners with the means of emphasizing specific force, velocity, and power outcomes. Different strategies should be considered in context of an individual’s needs. Systematic review registration: PROSPERO CRD42021259205.