PURPOSE: To investigate the physiological and performance effects of active and passive recovery between repeated-change-of-direction (RCOD) sprints. METHODS: Eight semi-professional basketball players (age: 19.9 ± 1.5 yr; stature: 183.0 ± 9.6 cm; body mass: 77.7 ± 16.9 kg; body fat: 11.8 ± 6.3%; VO2peak: 46.1 ± 7.6 mL·kg-1·min-1) completed 12 x 20-m RCOD sprints (Agility 5-0-5 Tests) interspersed with 20 s of active (50% maximal aerobic speed) or passive recovery in a randomized cross-over design. Physiological and perceptual measures included heart rate (HR), oxygen consumption (VO2), blood lactate concentration ([BLa-]), and rating of perceived exertion (RPE). Change-of-direction speed was measured during each sprint using the Change-of-Direction Deficit (CODD), with summed CODD time and CODD decrement calculated as performance measures. RESULTS: Average HR (7.3 ± 6.4 beats·min-1; p = 0.010; effect size (ES) = 1.09; very likely) and VO2 (4.4 ± 5.0 mL·kg-1·min-1; p = 0.12; ES = 0.77; unclear) were moderately greater with active recovery compared to passive recovery across sprints. Summed CODD time (0.87 ± 1.01 s; p = 0.07; ES = 0.76, moderate; likely) and CODD decrement (8.1 ± 3.7%; p < 0.01; ES = 1.94, large; almost certainly) were higher with active compared to passive recovery. Trivial-small differences were evident for RPE (p = 0.516, ES = 0.19, unclear) and post-test [BLa-] (p = 0.29, ES = 0.40, unclear) between recovery modes. CONCLUSIONS: Passive recovery between RCOD sprints may reduce the physiological stress and fatigue encountered compared to active recovery in basketball players.