The purpose of this investigation was to examine the effect of an 8-week training program with heavy- vs. light-load jump squats on various physical performance measures and electromyography (EMG). Twenty-six athletic men with varying levels of resistance training experience performed sessions of jump squats with either 30% (JS30, n = 9) or 80% (JS80, n = 10) of their one repetition maximum in the squat (1RM) or served as a control (C, n = 7). An agility test, 20-m sprint, and jump squats with 30% (30J), 55% (55J), and 80% (80J) of their 1RM were performed before and after training. Peak force, peak velocity (PV), peak power (PP), jump height, and average EMG (concentric phase) were calculated for the jumps. There were significant increases in PP and PV in the 30J, 55J, and 80J for the JS30 group (p < 0.05). The JS30 group also significantly increased in the 1RM with a trend towards improved 20-m sprint times. In contrast, the JS80 group significantly increased both PF and PP in the 55J and 80J and significantly increased in the 1RM but ran significantly slower in the 20-m sprint. In the 30J the JS30 group's percentage increase in EMG activity was significantly different from the C group. In the 80J the JS80 group's percentage increase in EMG activity was significantly different from the C group. This investigation indicates that training with light-load jump squats results in increased movement velocity capabilities and that velocity-specific changes in muscle activity may play a key role in this adaptation.