The aims of this study were to (1) investigate the influence of general anthropometric variables, handball-specific anthropometric variables, and upper-limb power and strength on ball-throwing velocity in a standing position (ν(ball)), and (2) predict this velocity using multiple regression methods. Forty-two skilled male handball players (age 21.0 ± 3.0 years; height = 1.81 ± 0.07 m; body mass = 78.3 ± 11.3 kg) participated in the study. We measured general anthropometric variables (height, body mass, lean mass, body mass index) and handball-specific anthropometric parameters (hand size, arm span). Upper-limb dynamic strength was assessed using a medicine ball (2 kg) throwing test, and power using a one-repetition maximum bench-press test. All the variables studied were correlated with ball velocity. Medicine ball throwing performance was the best predictor (r = 0.80). General anthropometric variables were better predictors (r = 0.55-0.70) than handball-specific anthropometric variables (r = 0.35-0.51). The best multiple regression model accounted for 74% of the total variance and included body mass, medicine ball throwing performance, and power output in the 20-kg bench press. The equation formulated could help trainers, athletes, and professionals detect future talent and test athletes' current fitness.