Statistical analyses were completed using JMP software (version 11; SAS Institute, Cary, NC, USA). Student’s t-tests were used to test if there were differences by position or seniority at preseason (p # 0.05 considered significant). Outcome measures were changes in weight and changes in whole body and regional lean and fat mass. Postseason changes (Post) were calculated as postseason minus preseason measure and indicate the changes over the competitive part of the season, whereas spring season changes (Spring) were spring season minus postseason measure and indicate changes during the off-season. The change at Post or Spring was considered significantly different from zero if the 95% confidence interval (CI) of the mean change in the outcome measure excluded zero. Regression analyses controlling for height and the baseline measure of the outcome variable were conducted to test if the changes at the Post or Spring time points were different by position, seniority, or the interaction between position and seniority. Baseline measure was defined as the preseason measure for Post outcomes, and the postseason measure for Spring outcomes. If the interaction between position and seniority was not significant, main effects of position and seniority groups were reported. A p value ,0.05 was considered significant for the interaction or for a difference between groups.