Equestrian rider trunk-pelvis strategies can be identified using self-organising maps

The purpose of the study was to explore whether equestrian riders can be grouped by their trunk-pelvis movement strategy. Riders (n = 40) with national or international competition experience in dressage were measured using motion capture on a riding simulator for ten seconds of simulated medium and extended trot, respectively. Trunk and pelvic pitch trajectories were filtered, time-normalised to the riding simulator’s vertical displacement cycle, and scaled. A self-organising map, with subsequent k-means clustering, identified three groups of rider-trunk pelvis movement. These groups related to the relative timing between peak posterior trunk and pelvis pitch. The study identified movement-based classifications of riders for future studies, which may have implications for rider injury risk and training.