The relationship between muscle size and strength has been published in the literature, however less is known about its actual association for adolescents. The identification of anthropometric muscle mass indicators that best predict strength across age taking in account growth and maturation processes, must be useful in field settings for screening sport talents, and sport training with young. PURPOSE: To analyze the association between lower-limb strength (LLS) and regional skeletal muscle mass (SMM) variables in young athletes and non-athletes. Secondly, to identify which regional SMM indicators are more successful predictors of LLS after adjusted for maturation. METHODS: SMM variables were assessed by anthropometric measures in 166 adolescents between 12 and 18 years old (15.2 ± 1.7 yrs, BMI = 21.9 ± 3.9 kg·m-2). The volunteers were recruited from local high-schools and local sport clubs. Athletes (AT) and non-athletes (NAT) were identified using an ad hoc questionnaire (NA = 99, 42 boys, 57 girls, and A = 67, 42 boys, 25 girls). Skeletal muscle mass (SMM) was calculated by anthropometric-derived equations. A dynamometer device was used to assess LLS. Partial correlations were carried out to determine associations between BC variables and strength values. The best-correlated variables with LLS performance were included in a stepwise regression analysis to determine SMM predictors. RESULTS: As expected AT were stronger than NA (80.3 ± 32.3 vs 58.4 ± 27.1 kg, p<0.001). The best correlation between SMM variables and LLS test was for total SMM in NAT boys after controlled by height (r = 0.520, p<0.001). When participants were pooled together, SMM, athlete condition and height were predictors of LLS test (R-squared = 0.484, p<0.001). Similar results were found for an alternative model including mid-calf corrected girth (CCG), athlete condition, gender and height (R-squared = 0.485, p<0.001). Surprisingly, age at peak height velocity was not found as a predictor of LLS. CONCLUSIONS: In accordance with other studies, our results confirm that lower-limb strength performance in adolescents was partially dependent on SMM. The main finding was that a simple measure as CCG can contribute to develop youth predictive models of sport performance and talent identification. Funded by Spanish Ministry of Economy and Competiveness (grant: DEP2011-30565)