In this study the effects of systematic manipulations of control and muscle strength on vertical jump height were investigated. Forward dynamic simulations of vertical squat jumps were performed with a model of the human musculoskeletal system. Model input was STIM(t), stimulation of six lower extremity muscles as function of time; model output was body motion. The model incorporated all features of the musculoskeletal system of human test subjects considered salient for vertical jumping, and the initial body configuration was set equal to that of the test subjects. First, optimal STIM(t) was found for a standard version of the model (experiment A). A satisfactory correspondence was found between simulation results and kinematics, kinetics and electromyograms of the test subjects. Subsequently, optimal STIM(t) for the standard model was used to drive a model with strengthened muscles (experiment B). Jump height was now lower than that found in experiment A. Finally, optimal STIM(t) was found for the model with strengthened muscles (experiment C). Jump height was now higher than that found in experiment A. These results suggest that in order to take full benefit of an increase in muscle strength, control needs to be adapted. It is speculated that in training programs aimed at improving jumping achievement, muscle training exercises should be accompanied by exercises that allow athletes to practice with their changed muscles.