The aim of our study was to compare the effects of two different plyometric training programs (targeting knee extensors or plantar flexors) on jump height and strength of leg muscles. Twenty‐nine male basketball players were assigned to the knee‐flexed (KF), knee‐extended (KE), or control groups. In addition to regular training, the KF group performed plyometric jumps (10 sets of 10 jumps, 3 sessions/week, 4 weeks) from 50 cm boxes with the knee flexed (90°–120°), whereas the KE group performed the jumps from 30 cm boxes with the knee much more extended (130°–170°). Jumping ability was evaluated with squat jumps (SJs), countermovement jumps (CMJs), and drop jumps from 20 cm (DJ20) and 40 cm (DJ40). Knee and ankle muscles were assessed during maximal isokinetic and isometric tests, and EMG activity was recorded from vastus lateralis and medial gastrocnemius. The KF group increased SJ (+10%, d = 0.86) and CMJ (+11%, d = 0.70) but decreased DJ40 height (−7%, d = −0.40). Conversely, the KE group increased DJ20 (+10%, d = 0.74) and DJ40 (+12%, d = 0.77) but decreased SJ height (−4%, d = −0.23). The reactivity index during DJs increased (+10% for DJ20, d = 0.47; +20% for DJ40, d = 0.91) for the KE group but decreased (−10%, d = −0.48) for the KF group during DJ40. Plantar flexor strength increased for the KE group (d = 0.72–1.00) but not for the KF group. Negative transfer across jumps is consistent with the principle of training specificity. Basketball players interested to perform fast rebounds in their training should avoid plyometric jumps with large knee flexions and long contact times.