Non linear anthropometric predictors in swimming.

In this paper we have tried to identify the significance and character of the linear and non-linear relations between simple anthropometric predictors: body height (BH), body weight (BW), and body mass index, and swimming performance: freestyle swimming 50 (FS50) and 400 meters (FS400), in a sample of young (15 years old on average) male (N = 40) and female (N = 28) swimmers. Linear (general model: y = a+bx) and nonlinear regression (general model: y = a+bx+cx2) were calculated simultaneously. Morphological variables are a significantly better predictor of the FS50 in males (BH mostly), and FS400 in females (BW mostly). This study emphasized some of the main advantages in the nonlinear regression calculation (including an interpretation of the relationships at a more superior level), and consequently allowed a precise anthropometric modeling in swimming using simple and easily measurable variables. For example, the best results in FS400 can be expected for the subjects that are average in BW (which guarantees solid muscle mass--the generator of force), but above average in BH (because of the physical law of lever). In conclusion, nonlinear regressions allow one to define the real nature of the relationships between variables, but only if compared with the linear ones. Additionally, this study emphasized one of the most important factors in defining possible specification-equation (e.g. structure of the influence of the different dimensions on the sport achievement) in different sports. In short, it underlines the importance of sampling the appropriate sample of the subject--highly skilled athletes exclusively.