Young soccer players are usually trained with adult-training methods, even though the physiological adaptations are likely to be very different compared with adults. In contrast, some have suggested training preadolescents only with coordination training. The purpose of this study was to investigate whether coordination or repeated-sprint training better improved speed over 20 m, with and without the ball. Sixteen soccer players (mean age 11 ± 0.5 y) were randomly assigned to a sprint-training group (STG = 7) or a coordination-training group (CTG = 9). The STG trained twice a week for 12 wk and performed 20 repetitions of 20- and 10-m sprints; the CTG performed coordination training (eg, speed ladder running) for the same training duration. Maximal jump height, anthropometric measures, and 20-m sprint time, with and without ball, were evaluated before and after the training period. Statistical significance was determined using two-way ANOVA with repeated measure and Pearson test for correlation. Both groups improved speed without the ball: STG = 3.75 ± 0.10 s to 3.66 ± 0.09 s (P < .05); CTG = 3.64 ± 0.13 s to 3.56 ± 0.13 s (P < .05), with no difference between groups. Sprint time with the ball pre- and posttraining was 4.06 ± 0.11 s and 4.05 ± 0.19 s (P > .05) for STG and 4.04 ± 0.12 s and 3.82 ± 0.15 s (P < .05) for CTG, with a significant difference between groups posttraining (P < .05). There were significant correlations between sprint time without ball, CMJ, and SJ. These data suggest that coordination training increases the speed with the ball more than typical repeated-sprint training. It can be hypothesized that running speed with ball improved more in CTG because this particular action requires improvements in coordination.