Identifying Key Physical Predictors of Sprint Skating Performance in Collegiate Hockey Athletes Using Regularized Regression Models

ABSTRACT: Fitzgerald, JS, Poolman, MA, and Secomb, JL. Identifying key physical predictors of sprint skating performance in collegiate hockey athletes using regularized regression models. J Strength Cond Res XX(X): 000-000, 2026-The purpose of this study was to identify the most influential lower-body physical capacities associated with the distinct sprint skating phases in collegiate hockey athletes. While numerous research has reported associations between various physical capacities and sprint skating performance, these have typically examined narrow subsets rather than a comprehensive profile, limiting practitioners' ability to determine the priority assessments in time-constrained high-performance environments. Twenty male NCAA Division I hockey athletes completed the following assessments: isometric hip strength (flexion, extension, adduction, abduction, and internal and external rotation), countermovement jump, single-leg lateral jump, broad jump, isometric belt squat, and 6-second cycling peak power. Thirty-five metres forward sprint skating performance was assessed from a stationary start and analyzed for initial acceleration (0-6 m), flying acceleration (10-20 m), and maximal performance (0-35 m) phases. Regularized regression analyses were performed to identify the most influential variables associated with each phase. Countermovement jump modified-reactive strength index was identified as the most influential and consistent predictor across all sprint skating phases. Hip external rotation relative peak force was also identified as a key predictor of initial acceleration, aligning with the mechanical importance of optimizing blade orientation and propulsive force generation. Hip flexion relative peak force contributed consistently across predictive models, highlighting its proposed role in efficient leg recovery. Other capacities contributed minimally or not at all to the predictive models. These results indicate that practitioners should consider prioritizing reactive strength measures and hip external rotation and flexion force assessments, when implementing profiling and monitoring protocols in high-performance hockey.