Diagnostics of performance prerequisites in sport in the context of the possible use of fuzzy theory
|Fakulta / Pracoviště MU
|The optimal level of the conditional abilities of an athlete is an important factor in tennis performance. The level of these abilities is generally diagnosed by motor tests and test batteries. The evaluation of these diagnostic tests is generally performed by means of a classical probabilistic approach. There are, however, also other mathematical theories that could be more suitable for this purpose, one of which is the fuzzy set theory, which (unlike the classical set theory) works with fuzzy borders between sets. The study put forward here aims to present possibilities for the use of the fuzzy approach in evaluating the results obtained by tennis players in the TENDIAG1 test battery and to compare the results obtained by tennis players in the TENDIAG1 test battery with the use of a probabilistic approach and a fuzzy approach. The TENDIAG1 test battery (9 items) contains somatic, fitness and coordination tests (each of these groups is comprised of three test). Somatic tests are informative (without points). The level of fitness and coordination tests was evaluated using three-level standards 0-2 points. The total score of the TENDIAG1 test battery was given by the sum of points obtained in the fitness and coordination tests (0–12 points). For fuzzy approach, S-shaped and Z-shaped membership functions with a linear function, a quadratic function and Gaussian function in the interval of the midlevel (midlevel is given by the interval (M - SD; M + SD) for each test item) were chosen. The research set was comprised of female tennis players aged 15–16 (n = 173, height: M ± SD = 169.93 ± 5.63 cm, weight: M ± SD = 61.29 ± 6.27 kg) who took part in regular testing by the Czech Tennis Federation in the years 2000–2017. No significant differences were found during comparison of the overall results of the evaluation using the probabilistic approach and the fuzzy approach with a linear (d = 0.45) or quadratic (d = 0.38) function in the midlevel. Significant differences were found (d = 1.12) during comparison of the overall results of the evaluation using the probabilistic approach and the fuzzy approach with a Gaussian function in the midlevel. In the context of the diagnosis and selection of tennis talent, the fuzzy approach appears to be more appropriate than the probabilistic approach as it allows for finer and more precise differentiation of the standard of the tested persons. The fuzzy approach also enables using various functions shapes, which makes it possible to take into account additional requirements on the evaluation in the midlevel interval. This could comprise the subject of further research.