Knowledge of wheelchair kinematics during a match is prerequisite for performance improvement in wheelchair basketball. Unfortunately, no measurement system providing key kinematic outcomes proved to be reliable in competition. In this study, the reliability of estimated wheelchair kinematics based on a three inertial measurement unit (IMU) configuration was assessed in wheelchair basketball match-like conditions. Twenty participants performed a series of tests reflecting different motion aspects of wheelchair basketball. During the tests wheelchair kinematics were simultaneously measured using IMUs on wheels and frame, and a 24-camera optical motion analysis system serving as gold standard. Results showed only small deviations of the IMU method compared to the gold standard, once a newly developed skid correction algorithm was applied. Calculated Root Mean Square Errors (RMSE) showed good estimates for frame displacement (RMSE ≤ 0.05 m) and speed (RMSE ≤ 0.1 m/s), except for three truly vigorous tests. Estimates of frame rotation in the horizontal plane (RMSE<3°) and rotational speed (RMSE < 7 ° /s) were very accurate. Differences in calculated Instantaneous Rotation Centres (IRC) were small, but somewhat larger in tests performed at high speed (RMSE up to 0.19 m). Average test outcomes for linear speed (ICCs > 0.90), rotational speed (ICC > 0.99) and IRC (ICC > 0.90) showed high correlations between IMU data and gold standard. IMU based estimation of wheelchair kinematics provided reliable results, except for brief moments of wheel skidding in truly vigorous tests. The IMU method is believed to enable prospective research in wheelchair basketball match conditions and contribute to individual support of athletes in everyday sports practice.