Normalization of Peak Torque in Adolescent Athletes: Which Method and What Body Parameters to Use?

This study compared models minimizing body-size effects on knee extension and flexion peak torque in adolescent athletes. The sample included 102 athletes (66 males, 13.9 years; 36 females, 14.2 years). Peak torque was measured at 60°/s with an isokinetic dynamometer. Body mass (BM), height, lean soft tissue (LST), and dominant-leg appendicular lean soft tissue (ALST) were assessed. Normalization required nonsignificant associations with body size (p > .05) and low explained variance (R2 < 9%). Cross-validation and AICc analysis were subsequently performed. Standard ratios did not achieve effective normalization simultaneously in both sexes. Seven allometric models met both criteria in both sexes, comprising single-variable models using LST, ALST, or maturity-adjusted LST, and combinations of BM and height, LST and height, ALST and height, or maturity-adjusted LST and height. LST-based models performed most consistently and are recommended, while ALST-based models and BM–height combinations remain viable alternatives.