Eye movements are usually studied under highly restricted and repeated conditions. To bridge the gap to natural sequential behavior while maintaining experimental control, we used a gamified interception task. Eye, head, and hand movements of participants playing an iPad-based game (Pong) were measured and gaze was mapped onto the tablet via computer vision algorithms. Participants attempted to hit a moving target (ball) by controlling a paddle on the lower part of the screen and scored points whenever an automated opponent missed the ball. We tested whether participants adapted their movements with respect to i) the impending action, ii) errors, after missing the target, and iii) prolonged exposure to experimental conditions, where paddle size and target speed were manipulated. Results showed adaptation of behavior on multiple levels: (1) Participants carefully timed their eye and hand movements with respect to the next action: They closely pursued the target with their gaze shortly before intercepting themselves. Blinks were clustered at times shortly after their own or the opponent’s action. (2) Participants initiated hand movements earlier after misses. Within trials and across participants, better performance in the game was associated with more pursuit eye movements and fewer head movements just before interception. (3) After long exposure, participants pursued the opponent more often compared to the beginning of the experiment, although overall behavioral patterns stayed roughly consistent. Together, these results indicate flexible adaptation of eye, head, and hand movements for a given situation, but stable behavioral patterns across prolonged exposure which are related to performance.