A novel validation approach shows new, solid reasons why vertical jump height should not be used to predict leg power

Abstract

Jump height continues to bewidely used to predict power in humans. Individual progress is often monitored on the basis of estimated power, butprediction equations are based on group data. The objective of the study wasto show that vertical jump performance (VJP) and mechanical power are poorly associated, particularly within individuals.Two experiments are presented. First, 52 physically active male college students performed five maximal vertical jumps each. Second, three young male participantsperformed 50 maximal jumps each. Participants rested for 1 minute between jumps. VJP was calculated from kinematic data as peak body center of mass (BCOM) minus standing BCOM; peak power (PEAKPWR) was calculated from the vertical ground reaction force registered by a force plate, and average power(MEANPWR) during propulsion from the change in potential energy of BCOM. Regression analyses were performed usingstandardized VJP scores as the predictor variable and standardized powerscores as the resulting variables,expecting an identity function of y = x (intercept = 0, slope = 1)and R2= 1. In experiment 1,themodelforzPEAKPWR R2= 0.9707 (p < 0.0001) but slope (0.3452) ≠ 1 (p < 0.0001). The modelforzMEANPWRR2= 0.9239 (p < 0.0001); nevertheless, slope (0.4257) ≠ 1 (p< 0.0001).In experiment 2, allindividual models for zPEAKPWR and zMEANPWR resulted in poor associations (R2≤0.21) and slopes ≠ 1 (p≤0.001). In conclusion, regression analysis for individuals, and even for groups,confirmsthat VJP is a poor predictor of mechanical power.

Keywords

• kinematics
• biomechanical phenomena
• biomechanics
• sports
• lower limbs
• validation
• within-subject analysis