Teixeira José E, Encarnação Samuel, Branquinho Luís, Morgans Ryland, Afonso Pedro, Rocha João, Graça Francisco, Barbosa Tiago M, Monteiro António M, Ferraz Ricardo, Forte Pedro
Department of Sport Sciences, Polytechnic of Guarda, 6300-559 Guarda, Portugal.
Department of Sport Sciences, Instituto Politécnico de Bragança, 5300-253 Bragança, Portugal.
J Funct Morphol Kinesiol. 2024 Jun 28;9(3):114. doi: 10.3390/jfmk9030114.
The aim of this study was to test a machine learning (ML) model to predict high-intensity actions and body impacts during youth football training. Sixty under-15, -17, and -19 sub-elite Portuguese football players were monitored over a 6-week period. External training load data were collected from the target variables of accelerations (ACCs), decelerations (DECs), and dynamic stress load (DSL) using an 18 Hz global positioning system (GPS). Additionally, we monitored the perceived exertion and biological characteristics using total quality recovery (TQR), rating of perceived exertion (RPE), session RPE (sRPE), chronological age, maturation offset (MO), and age at peak height velocity (APHV). The ML model was computed by a feature selection process with a linear regression forecast and bootstrap method. The predictive analysis revealed that the players' MO demonstrated varying degrees of effectiveness in predicting their DEC and ACC across different ranges of IQR. After predictive analysis, the following performance values were observed: DEC (x¯ = 41, β = 3.24, intercept = 37.0), lower IQR (IQR = 36.6, β = 3.24, intercept = 37.0), and upper IQR (IQR = 46 decelerations, β = 3.24, intercept = 37.0). The player's MO also demonstrated the ability to predict their upper IQR (IQR = 51, β = 3.8, intercept = 40.62), lower IQR (IQR = 40, β = 3.8, intercept = 40.62), and ACC (x¯ = 46 accelerations, β = 3.8, intercept = 40.62). The ML model showed poor performance in predicting the players' ACC and DEC using MO (MSE = 2.47-4.76; RMSE = 1.57-2.18: R = -0.78-0.02). Maturational concerns are prevalent in football performance and should be regularly checked, as the current ML model treated MO as the sole variable for ACC, DEC, and DSL. Applying ML models to assess automated tracking data can be an effective strategy, particularly in the context of forecasting peak ACC, DEC, and bodily effects in sub-elite youth football training.
本研究的目的是测试一种机器学习（ML）模型，以预测青少年足球训练期间的高强度动作和身体碰撞。在为期6周的时间里，对60名15岁、17岁和19岁以下的葡萄牙次精英足球运动员进行了监测。使用18赫兹的全球定位系统（GPS）从加速度（ACC）、减速度（DEC）和动态应力负荷（DSL）等目标变量中收集外部训练负荷数据。此外，我们使用总质量恢复（TQR）、主观用力程度分级（RPE）、训练课主观用力程度分级（sRPE）、实际年龄、成熟偏移（MO）和身高增长峰值年龄（APHV）来监测主观用力程度和生物学特征。ML模型通过具有线性回归预测和自助法的特征选择过程进行计算。预测分析表明，球员的MO在预测不同四分位间距范围内的DEC和ACC时表现出不同程度的有效性。经过预测分析，观察到以下性能值：DEC（x¯ = 41，β = 3.24，截距 = 37.0），下四分位间距（IQR = 36.6，β = 3.24，截距 = 37.0），上四分位间距（IQR = 46次减速度，β = 3.24，截距 = 37.0）。球员的MO还表现出预测其上四分位间距（IQR = 51，β = 3.8，截距 = 40.62）、下四分位间距（IQR = 40，β = 3.8，截距 = 40.62）和ACC（x¯ = 46次加速度，β = 3.8，截距 = 40.62）的能力。ML模型在使用MO预测球员的ACC和DEC方面表现不佳（均方误差 = 2.47 - 4.76；均方根误差 = 1.57 - 2.18：R = -0.78 - 0.02）。成熟度问题在足球表现中普遍存在，应定期检查，因为当前的ML模型将MO视为ACC、DEC和DSL的唯一变量。应用ML模型评估自动跟踪数据可能是一种有效的策略，特别是在预测次精英青少年足球训练中的峰值ACC、DEC和身体影响的背景下。
Zotero 插件