Objective of the study is to develop a mathematical model for predicting the maximum oxygen consumption (MOC) of football players aged 15-17 based on anthropometric and functional indicators. Methods and structure of the study. To build a prognostic model, a set of anthropometric indicators (body length and weight, thickness of skin and fat folds, body composition), heart rate variability parameters (HR, SI, TP, HF, LF, IC, PARS), as well as (systolic and diastolic blood pressure) were preliminarily determined. The MOC was measured by the direct method when performing a maximum load test using a gas analyzer. Five machine learning algorithms have been tested. The developed mathematical model based on linear regression makes it possible to predict the MOC of football players with a fairly high degree of accuracy. Results and conclusions. During development, five different machine learning algorithms were tested to predict MOC in football players based on a set of anthropometric and functional indicators, and a comparative analysis of the algorithms was performed. A mathematical model based on the use of a linear regression algorithm allows the maximum oxygen consumption of football players to be predicted with high accuracy (R²=0.78; MAE=3.1 ml/kg/min), which meets the criteria for prediction in sports physiology and can be used for practical application in the training process.
© Copyright 2025 Theory and Practice of Physical Culture. ANO SPC "Theory and Practice of Physical Culture and Sport". All rights reserved.