Purpose: To examine the physiological, physical, and technical demands of game-based drills (GBDs) with regular dribble (RD) or no dribble (ND) involving a different number of players (3 vs 3, 4 vs 4, and 5 vs 5). Methods: Ten regional-level male basketball players performed 6 full-court GBD formats (each consisting of 3 bouts of 4 min and 2 min rest) on multiple occasions. The physiological and perceptual responses were measured through heart rate and rating of perceived exertion. Video-based time–motion analysis was performed to assess the GBD physical demands. The frequencies of occurrence and the duration were calculated for high-intensity, moderate-intensity, low-intensity, and recovery activities. Technical demands were assessed with a notional-analysis technique. A 2-way repeated-measures analysis of variance was used to assess statistical differences between GBD formats. Results: A greater perceptual response (rating of perceived exertion) was recorded during 3 versus 3 than 5 versus 5 formats (P = .005). Significant interactions were observed for the number of recovery (P = .021), low-intensity activity (P = .007), and all movements (P = .001) completed. Greater time was spent performing low-intensity and high-intensity activities during RD than ND format. Greater technical demands were observed for several variables during 3 versus 3 than 4 versus 4 or 5 versus 5. A greater number of turnovers (P = .027), total (P ≤ .001), and correct passes (P ≤ .001) were recorded during ND than RD format. Conclusions: The number of players predominantly affected the perceptual response to GBD, while both the number of players and rule modification (RD vs ND) affected activities performed during GBD. Reducing the number of players increases the GBD technical elements, while ND format promotes a greater number of turnovers and passes.