The aim of this study is to investigate the effect of virtual reality (VR)-based training on the balance and agility performance of amateur boxers. The study was conducted using a pre-test-post-test control group experimental design. A total of 26 male athletes (13 experimental, 13 control) who were boxing in Mardin province, aged 18–26, with at least two years of licensed sports experience were randomly selected and included in the study. The experimental group underwent VR-based training for 8 weeks, 3 days a week, for an average of 10 minutes after each training session, using the Meta Quest 3 virtual reality headset and The Thrill of the Fight VR Boxing software. The control group continued with their traditional training regimen. Performance evaluation was conducted using the T agility test and the Stork balance test (dominant and non-dominant leg). Data were analyzed using SPSS 26.0, and the significance level was set at p<0.05 using multivariate analysis of variance (MANOVA). A significant increase was observed in the T agility test in the experimental group (p<0.001), and the time × group interaction was found to be statistically significant (p=0.011). In the Stork test, the time effect was significant in the dominant leg, but the group and interaction effects were not significant. In the Stork test, no significant differences were found in any of the effects in the non-dominant leg. SG-based training programs have positive and significant effects, particularly on dynamic motor skills such as agility. In terms of balance development, the effects are more limited and variable, and may differ depending on the type of test applied. Virtual reality systems have potential as a complementary tool in sports training and should be supported by longer-term, sport-specific protocols in the future.