The aims of the present study were to compare the effects of 1) training at 90 and 100% sprint velocity and 2) supervised versus unsupervised sprint training on soccer-specific physical performance in junior soccer players. Young, male soccer players (17 ± 1 yr, 71 ± 10 kg, 180 ± 6 cm) were randomly assigned to four different treatment conditions over a 7-week intervention period. A control group (CON, n = 9) completed regular soccer training according to their teams' original training plans. Three training groups performed a weekly repeated-sprint training session in addition to their regular soccer training sessions performed at A) 100% intensity without supervision (100UNSUP, n = 13), B) 90% of maximal sprint velocity with supervision (90SUP, n = 10) or C) 90% of maximal sprint velocity without supervision (90UNSUP, n=13). Repetitions x distance for the sprint-training sessions were 15 x 20 m for 100UNSUP and 30 x 20 m for 90SUP and 90UNSUP. Single-sprint performance (best time from 15 x 20 m sprints), repeated-sprint performance (mean time over 15 x 20 m sprints), countermovement jump and Yo-Yo Intermittent Recovery Level 1 (Yo-Yo IR1) were assessed during pre-training and post-training tests. No significant differences in performance outcomes were observed across groups. 90SUP improved Yo-Yo IR1 by a moderate margin compared to controls, while all other effect magnitudes were trivial or small. In conclusion, neither weekly sprint training at 90 or 100% velocity, nor supervised sprint training enhanced soccer-specific physical performance in junior soccer players.