While self-controlled practice has been shown to enhance motor learning with various populations and novel tasks, it remains unclear if such effects would be found with athletes completing familiar tasks. Study 1 used a single case-study design with a world-champion kickboxer. We investigated whether giving the athlete a choice over the order of punches would affect punching velocity and impact force. Separated by 1 min of rest, the athlete completed 2 rounds of 12 single, maximal effort punches (lead straight, rear straight, lead hook and rear hook) delivered to a punching integrator in a counterbalanced order over six testing days. In one round the punches were delivered in a predetermined order while in the second round the order was self-selected by the athlete. In the choice condition, the world champion punched with greater velocities (6–11 %) and impact forces (5–10 %). In Study 2, the same testing procedures were repeated with 13 amateur male kickboxers over 2 testing days. Similar to Study 1, the athletes punched with significantly greater velocities (6 %, p < 0.05) and normalised impact forces (2 %, p < 0.05) in the choice condition. These findings complement research on autonomy support in motor learning by demonstrating immediate advantages in force production and velocity with experienced athletes.