A machine learning approach for the classification of sports based on a coaches' perspective of environmental, individual and task requirements: A sports profile analysis.

Well-designed talent programmes in sports with a focus on talent identification, orientation, development, and transfer support the engagement of young individuals and the pursuit of elite performance. To facilitate these processes, an analysis of task, environmental and individual characteristics per sport is much needed. The aims of this study were to 1) analyse whether unique profiles per sport could be established by generic characteristics and 2) to discuss similarities and differences for the potential application in talent development and transfer. By means of a validated survey, 1247 coaches from 34 sports ranked 18 characteristics on importance to their sports (0 = not important - 10 = very important). To discriminate the responses per sport a Discriminant Analysis (DA) was carried out. To refine the DA-classification, Uniform Manifold Approximation and Projection (UMAP) with CatBoost classifier was performed. To test the performance of the CatBoost classifier-algorithm, a confusion-matrix was generated. The cross-validated DA showed that 70.2% of the coaches were correctly classified to their sport. The UMAP/CatBoost technique revealed 75.1% accuracy with correctly predicted responses per sport ranging from 18.2% (sailing) to 98.2% (soccer). With varying precision, the algorithm was able to differentiate sports by importance of its characteristics indicating similarities and differences per sport.

Factors affecting sports involvement in a school-based youth cohort: Implications for long-term athletic development.

The aim of the present study was to determine the factors affecting sports involvement in a school-based adolescent population. The cross-sectional cohort study assessed anthropometry, physical capacities and motor competence in 501 boys (aged 10-16 y), from junior (10-12 y) and senior (13-16 y) cohorts. Sports participation data was collected from junior participants. Multivariate analysis of covariance revealed moderate maturity, anthropometry, physical capacity and motor competence differences between sports in the senior cohort (F = 2.616, p < 0.001, η2 = .08), but not in the junior cohort. Furthermore, differences in physical fitness were revealed between playing levels (F = 2.616, p < 0.001, η2 = .08), with a discriminant analysis correctly classifying 73% of participants using aerobic fitness and vertical jump measures. Representative level participants engaged in more structured training and commenced organised competition at a later age (F = 4.332, p < 0.001, η2 = .21). This study's findings are twofold: 1) physical and motor competence profiles differ more between sports with increasing age, and 2) participants at a higher level of competition report delayed engagement in their main sport. As a result, schools may be the ideal environment in which to provide children and adolescents with the opportunity to sample different sports.

Generic anthropometric and performance characteristics among elite adolescent boys in nine different sports.

The aim of the present study was to evaluate the Flemish Sports Compass (FSC), a non-sport-specific generic testing battery. It was hypothesised that a set of 22 tests would have sufficient discriminant power to allocate athletes to their own sport based on a unique combination of test scores. First, discriminant analyses were applied to the 22 tests of anthropometry, physical fitness and motor coordination in 141 boys under age 18 (16.1 ± 0.8 years) and post age at peak height velocity (maturity offset = 2.674 ± 0.926) from Flemish Top Sport Academies for badminton, basketball, gymnastics, handball, judo, soccer, table tennis, triathlon and volleyball. Second, nine sequential discriminant analyses were used to assess the ability of a set of relevant performance characteristics classifying participants and non-participants for the respective sports. Discriminant analyses resulted in a 96.4% correct classification of all participants for the nine different sports. When focusing on relevant performance characteristics, 80.1% to 97.2% of the total test sample was classified correctly within their respective disciplines. The discriminating characteristics were briefly the following: flexibility in gymnastics, explosive lower-limb strength in badminton and volleyball, speed and agility in badminton, judo, soccer and volleyball, upper-body strength in badminton, basketball and gymnastics, cardiorespiratory endurance in triathletes, dribbling skills in handball, basketball and soccer and overhead-throwing skills in badminton and volleyball. The generic talent characteristics of the FSC enable the distinction of adolescent boys according to their particular sport. Implications for talent programmes are discussed.