The sprint skill is predominant in wheelchair basketball game strategy [Bloxham, 2001] and need to be optimized during the training session. The development of sensors fixed to the wheelchair gives complementary and objective data for the coach and players [van der Slikke, 2017]. The aim of the study was to define a propulsive efficiency index with wearable sensor during a sprint session. Eleven wheelchair basketball players (32.5 ± 2.5 years, 70.1 ± 22.3 kg, IFWB classification 2.9 ± 1.6 points) realized 2*10 m sprint trials in a straight line on taraflex floor with their basketball wheelchair. The angular velocity of the right wheel is measured by the inertial sensor fixed to the hub (50 Hz, WheelPerf, AtoutNovation). A modelization of the min and max peaks of velocities for each push was defined by the computation of a 2-degree polynomial function. The propulsive efficiency is associated with the quantification of the R squared of the polynomial function. The maximum velocity (Vmax) and its location during the sprint were calculated. The times between the 2 trials are similar (CV = 1.3%). Although Vmax was reached mainly at the end of the sprint (83 to 100%) and required between 8 and 10 pushes, Vmax is different between the players (3.4 to 4.7 m.s −1 ). The value of the R square is different between the players (0.91 to 0.64) but similar for each of the two trials (0.02). The variations in the peaks velocity distribution induce some elements to improve the propulsion method according to the classifications and the adjustments of the wheelchair. The use a wearable sensor to quantify motor efficiency seems to be a relevant tool for the assessment and monitoring of wheelchair basketball players.