Predicting Linear Skating Speed by College Ice Hockey Players

Background/Purpose: As sport becomes increasingly science-based, coaches and scientists have designed sophisticated means of evaluating and predicting athlete performance.  Such information enables coaches to make sound decisions relative to player selection, position assignment and physical performance.  The purpose of this study was to examine dry-land performance tests as predictors of linear skating speed by male collegiate ice hockey players.  It was hypothesized that dry-land tests incorporating horizontal acceleration movements would be better predictors of linear skating speed than tests employing vertical explosive power movements.

Method: Forty Division I male college ice hockey players participated.  The design was correlational.  Players were individually tested on several parameters related to on-ice performance.  Independent variables (dry land) included the vertical jump (VJ), standing broad jump (SBJ), 40-yard sprint (40YD), and one repetition maximal back squat (1RMBS).  Dependent variables (skating speed) included the forward 90-ft acceleration test (AF), the backward 90-ft acceleration test (AB) and the flying 50-ft top speed test (F50).  All testing was conducted by qualified strength/speed/ conditioning specialists.

Analysis/Results: To determine the relative contribution of each independent variable, three separate multiple regression equations were run; one for each dependent variable.  Analyses indicated that only the vertical jump test had a significant association with any of the dependent variables (skating speed tests), AF: b= -.029, t (35)= -2.35, p= .025; AB: b= -.055, t (35) = -2.41, p=. 021; F50: b= - .029, t (35) = -2.68, p=. 011.  In other words, as vertical jump distance increased, skating times decreased. 

Conclusions: Contrary to the hypothesized outcome, only the vertical jump test had meaningful associations with three tests of linear skating speed.  It is possible that, due to players' inefficient off-ice running and forward jumping mechanics, the 40YD and SBJ were not valid predictors of linear skating speed.  Otherwise, results might help coaches avoid redundant testing and improve player evaluation.  Further studies should closely examine biomechanical relationships between the vertical jump test and linear skating speed.