Purpose: This study aimed to compare the effects of four resistance training (RT) programs with different velocity loss (VL) thresholds: 0% (VL0), 10% (VL10), 20% (VL20) and 40% (VL40) on sprint and jump performance, muscle strength, neuromuscular, muscle hypertrophy and architectural adaptations. Methods: Sixty-four young resistance-trained men were randomly assigned into four groups (VL0, VL10, VL20, and VL40) that differed in the VL allowed in each set. Subjects followed an RT program for 8 weeks (2 sessions per week) using the full-squat (SQ) exercise, with similar relative intensity (70-85% 1RM), number of sets (3) and inter-set recovery period (4-min). Before and after the RT program the following tests were performed: 1) muscle hypertrophy and architecture of the vastus lateralis (VLA); 2) tensiomyography; 3) 20-m running sprint; 4) vertical jump; 5) maximal voluntary isometric contraction in SQ; 6) progressive loading test in SQ; and 7) fatigue test. Results: No between-group differences existed for RT-induced gains in sprint, jump and strength performance despite the differences in the total volume performed by each group. VL20 and VL40 showed significant increases (P < 0.001) in muscle hypertrophy (group × time interaction, P = 0.06). However, only VL40 exhibited a significant slowing (P < 0.001) of the delay time in the VLA muscle (group × time interaction: P = 0.05). Moreover, VL40 showed a significant decrease in the early rate of force development (P = 0.04). Conclusions: Higher VL thresholds (i.e. VL20 and VL40) maximized hypertrophic adaptations, although an excessive VL during the set (i.e. VL40) may also induce negative neuromuscular adaptations. Therefore, moderate VL thresholds should be chosen to maximize strength adaptations and to prevent negative neuromuscular adaptations.