RÉSUMÉ
PURPOSE: To examine the longitudinal relationships between shoulder internal and external rotation (IR and ER) strength, maturity status, and swim performance (aim 1). To determine whether maturity status mediated (partially/fully) the relationship between shoulder IR/ER strength and performance in age-group swimmers (aim 2). METHODS: Using a repeated-measures design, anthropometrics, maturity status, shoulder IR/ER strength, and 200-m front-crawl velocity were assessed over 3 competition seasons in N = 82 Australian male competitive swimmers (10-15 y). For aim 1, linear mixed models examined longitudinal relationships between assessed variables. For aim 2, causal mediation analyses examined proportional (in)direct contributions of maturity status between shoulder IR strength and swim performance. RESULTS: For aim 1, linear mixed models identified a significant relationship between shoulder IR strength and swim performance over time (F1,341.25 = 16.66, P < .001, marginal R2 = .13, conditional R2 = .91). However, maturity status was influential (ΔAkaike information criterion = -75.8, χ2 = 19.98, P < .001), suggesting removal of the shoulder IR strength-swim velocity relationship (F1,214.1 = 0.02, P = .88). For aim 2, mediation analyses identified maturity status as fully mediating the shoulder IR strength-swim velocity relationship (92.30%, P < .001). CONCLUSIONS: Shoulder IR and ER strength did not account for variance in longitudinal age-group swim performance independent of maturity status. Interindividual differences in maturity status fully explained the relationship between shoulder IR/ER strength and swim performance. For practitioners, findings promote the need to account for maturation status and question the rationale for upper-limb strength assessment during maturational years.