Plyometric training (PT) programs are widely used to improve explosive actions in soccer players of various ages, although there is debate about optimal training duration and time course of improvement. Twenty-two early to mid-puberty elite soccer players were assigned to a control group (CG, n = 10, regular soccer training) or a plyometric training group (PTG, n = 12, regular soccer training substituted with 2 PT sessions each week). Both groups trained for 16 weeks during the in-season period. Control group performed only tests at baseline and after intervention, whereas PTG performed additional tests after 4, 8, and 12 weeks. During each test, subjects' performances in speed (10 and 30 m; 5 and 20 m), agility, shuttle run, multiple 5 bounds (MB5), and standing long jump (LJ) were recorded. The PTG showed improved performance in 20-m sprint time (-3.2%), agility time (-6.1%), MB5 distance (+11.8%), and LJ distance (+7.3%) (all, p ≤ 0.05) after 16 weeks. All these improvements were higher compared with CG (all, p ≤ 0.05). The time course of improvement in the PT group showed that 20-m sprint time improved after 16 weeks (p = 0.012); agility after 4 (p = 0.047) and 8 weeks (p = 0.004) but stopped after 12 weeks (p = 0.007); MB5 after 8 (p = 0.039), 12 (p = 0.028), and 16 weeks (p < 0.001); and LJ improved after 4 (p = 0.045), 12 (p = 0.008), and 16 weeks (p < 0.001). Plyometric training seems to be an appropriate training tool to enhance some but not all explosive actions. The results indicate that the duration of a PT program is highly dependent on what type of explosive actions should be improved, or whether several explosive actions should be improved at the same time.