The purpose of this study was to test which specific type of exercise (i.e., jump squat (JS) or half-squat (HS)) is more effective at maintaining speedand power abilities throughout a preseason in soccer players. Twenty-three male soccer players were randomly allocated into two groups: JS and HS. The mean propulsive power, vertical jumping ability, and sprinting performance were evaluated before and after 4 weeks of a preseason period. The optimum power loads for the JS and HS exercises were assessed and were used as load-references. The soccer players performed 10 poweroriented training sessions in total. Both JS and HS maintained power in JS and speed abilities (P > 0.05, for main effects and interaction effect) as indicated by ANCOVA. Both groups demonstrated reduced power during HS (ES = -0.76 vs. -0.78, for JS and HS, respectively); both groups improved acceleration (ACC) from 5 to 10 m (ES = 0.52). JS was more effective at reducing the ACC decrements over 0-5 m (ES = -0.38 vs. -0.58, for JS and HS, respectively). The HS group increased squat jump height (ES = 0.76 vs. 0.11, for HS and JS, respectively). In summary, JS is more effective in reducing the ACC capacity over very short sprints while HS is more effective in improving squat jump performance. Both strategies improve ACC over longer distances. New training strategies should be implemented/developed to avoid concurrent training effects between powerand endurance adaptations during professional soccer preseasons.