Use of principal component analysis for identification of differences in countermovement jump metrics between positions in collegiate baseball

Abstract

The countermovement jump (CMJ) has previously been implemented in collegiate baseball for assessing lower body neuromuscular characteristics and monitoring the accumulation of fatigue. However, it remains unclear as to which metrics are the most relevant for each position. The purpose of this investigation was to evaluate differences in CMJ metrics between positions in a cohort of collegiate baseball players using dimensional reduction techniques. Forty-five collegiate baseball players participated in the current investigation. All athletes performed two maximal effort CMJs on a dual uniaxial force plate system. A one-way analysis of variance (ANOVA) was used to detect statistically significant differences (p < 0.05) in CMJ metrics between positions, with participants classified as Catchers (CATs), Infielders (IFs), Outfielders (OFs), and Pitchers (PITs). Principal component analysis (PCA) and linear (LDA) and quadratic discriminant analysis (QDA) were used to assist in identification of most relevant CMJ metrics for each position. ANOVA detected significant differences between PITs and IFs for eccentric mean power (p = 0.048) and eccentric peak velocity (p = 0.040). PCA revealed that contraction time, impulse, and the duration of the braking, eccentric, and concentric phases had the highest contributions to PC1 and PC2. LDA and QDA identified metrics associated with the eccentric phase as explaining most of the variance between positions. The differences observed in eccentric CMJ metrics for PITs may be due to the greater eccentric loading that is experienced by the lower limbs during the different phases of a pitch. Future investigations are need for determine sensitivity of eccentric metrics of the CMJ to varying levels of fatigue.