This study compared the effects of 10% and 30% velocity-loss thresholds during Smith-machine squat training on maximal strength, jumping, and sprint performance in trained female collegiate athletes. Twenty-eight athletes were randomly assigned to a 10% velocity-loss group (VL10) or a 30% velocity-loss group (VL30).Both groups completed 8 weeks of supervised squat training twice weekly, performing four sets at 70% of one-repetition maximum, with each set terminated when the assigned threshold was reached. Smith-machine squat one-repetition maximum (1RM), squat jump (SJ), countermovement jump (CMJ), standing long jump (SLJ), and 30-m sprint (T30) performance were assessed before and after the intervention. Group × time effects were examined using mixed-design analyses of variance. Mean propulsive velocity was higher in VL10, whereas VL30 completed substantially more repetitions per set and per session (all p < 0.001). All performance outcomes improved over time (all p < 0.001). VL30 produced a greater increase in squat 1RM than VL10 (12.5% vs. 9.2%; interaction p < 0.001, ηp² = 0.450). VL10 produced greater improvements in SLJ (interaction p = 0.031, ηp² = 0.166) and T30 time (interaction p = 0.006, ηp² = 0.254). Group × time interactions were not significant for SJ (p = 0.127) or CMJ (p = 0.058). Both thresholds improved strength and lower-body performance, but the adaptation pattern differed by outcome. Under the present protocol, VL30 favored 1RM squat strength, whereas VL10 favored SLJ and T30 performance.