A Principal Component Analysis Comparing Forward Skating Strides Pre-Post Skate Treadmill Training in Youth Hockey Players

Abstract

The purpose of the study was to explore temporal and phasic waveform patterns within kinematic data, obtained from forward skating strides, pre-post skate treadmill training in youth (U-11), male hockey players to investigate changes in kinematics. Continuous joint angle (deg) and angular velocity (deg/s) stride data for the trunk, hip, and knee were determined, time normalized, and averaged. PCA results suggested that most of the pre-post variance in skating mechanics could be explained through an increase in joint angle (deg) and angular velocity (deg/s) magnitudes during the propulsive and recovery phases of the stride cycle. Single component reconstruction (SCR) facilitated visual representation and interpretation of kinematic differences by isolating variances within each principal component and reconstructing lower (5th percentile) and upper (95th percentile) waveforms based on the respective scalar weight factor of PC scores. Post-training, SCR suggested patterns of increased trunk extension throughout the stride cycle, increased hip and knee extension during propulsion, and increased hip and knee flexion during recovery. Statistical parametric mapping (SPM) statistically analysed continuous data points across two waveforms. SPM revealed significant differences in pre-post trunk flexion and extension magnitudes from the early glide phase to propulsion onset ([p=0.0023], ~0-20% stride cycle) and from late propulsion to weight acceptance ([p=0.0001], ~40-80% stride cycle). Differences in pre-post hip and knee measures were non-significant (p>0.05). PCA, SCR and SPM analyses have the potential to contribute to our understanding of biomechanical training adaptations in stride mechanics in youth ice hockey players.