Prediction of Handball Players' Performance on the Basis of Kinanthropometric Variables, Conditioning Abilities, and Handball Skills.

Handball (team handball) is a multifactorial sport. The aims of this study were (i) to analyse anthropometric variables, conditioning abilities, and handball skills in club handball players according to age and sex, and (ii) to develop multivariate models explaining club handball performance from a multidimensional perspective. Two hundred and twenty six handball players (age 16.9 ± 4.0 years, 54% males) participated in the study. The players belonged to under-14, under-16, under-19, and A teams. They were evaluated with a battery of 18 tests covering kinanthropometry, conditioning abilities, and handball skills. A one-way ANOVA with a Bonferroni post-hoc test was used to investigate differences between teams, and a t-test for differences between the sexes. For each team, a discriminant analysis was performed to determine differences between performance levels. The results showed little differences between the U19 and A teams in any of the variables studied in either men or women, and that the lowest values corresponded to the U14 team. The differences according to sex were clear in the kinanthropometric and conditioning variables, but much less so in handball skills. The eight multivariate models that were constructed classified successfully from 48.5 to 100% of the sample using at most three variables (except for the women's A team whose model selected six variables). Conditioning variables were most discriminating in men, and handball skills in women. This would seem to reflect the different performance profiles.