One hundred and ninety-four elite athletes performed jump squats against loads corresponding to 40%, 60%, and 80% of their respective body-mass. A linear regression analysis was performed to establish the relationship between muscle power expressions and jump squat height.
The coefficient of determination (R2) in the different linear regression models between muscle power-related variables and jump squat height, for the different load ranges, varied from 0.50 to 0.57 (for absolute power values) and from 0.72 to 0.78 (for relative power values [W/kg]). The mean propulsive power presented similar capacity to predict the jump squat height as the peakpower-related values. For all analyzed variables, this prediction power was increased when the absolute power values were normalized by the individuals' body-mass.
Selection of the values related to the mean propulsive phase to assess top-level athletes might be considered as an advantageous alternative, due to its adequacy to properly reflect the neuromuscular potential of the subjects in both ballistic and traditional exercises.