Power Output Prediction From Jump Height and Body Mass Does Not Appropriately Categorize or Rank Athletes

The purposes of this study were (a) to verify the agreement of categorization and ranks based on the actual power output measured by a force plate (PPact) and the estimated power output (PPest) from jump height and body mass (BM), and (b) to verify whether the ratio standard is adequate to scale the PPact for BM. The countermovement jumps of 309 male athletes were analyzed. The athletes were first categorized into tertiles (superior, intermediate, and inferior) according to PPact and PPest. After that the athletes were ranked (highest to lowest power output) according to PPact and PPest. The PPest equation explained 81% of PPact variance (standard error of estimate = 277.4 W). The PPest (3,757.1 ± 579.8 W) displayed similar mean values compared with PPact (3,757.1 ± 642.3 W). However, the agreement between the categories generated by PPact and PPest was only moderate (k = 0.6; p < 0.01), and in the intermediate tertile, the categorization differs 38.8%. The agreement between the ranks analyzed from a Bland-Altman plot shows bias zero, but a wide limits of agreement (81 ranks; 26.2%). For the PPact scaling, the ratio standard may be considered as an adequate method for removing the BM effect, considering the lack of correlation between the scaled PPact (PPact/BM) and BM, and also the confirmation of Tanner's special circumstance. In conclusion, our findings indicate that the athlete's power output was not appropriately categorized or ranked when using PPest. Furthermore, the use of the scaled PPact is recommended to fairly compare athletes with different BMs.