Methods / Statistics

Document Type



To account for athletes of different sizes, kinetic values are commonly ‘normalized’ by dividing by mass and/or height. However, the creation of a ratio variable requires certain statistical assumptions to be met. The purpose of this study was to determine if elbow valgus torque predicted by pitching velocity is influenced by the normalization method using regression model comparison with normalized torque values. Both mass and mass*height normalization satisfied the correlation and zero intercept assumptions. Results did not agree between analysis methods that elbow valgus torque could be predicted with pitching velocity at the α = 0.05 level, indicating caution should be exercised before normalizing pitching kinetics data without confirming the assumptions for a ratio variable are met.