Scholarlink

<p>This study analyzes performance data from training sessions of male soccer players to predict their match performance. Regression analysis was conducted to examine the relationship between training metrics—such as energy expenditure, sprint count, and power plays—and match performance features. Additionally, a binary classification model was employed to determine whether a player’s match performance could be predicted based on their training sessions leading up to the game. A correlation analysis using multivariate regression further explored the relationship between training session data and match outcomes.</p><p>To support training optimization, a drill selection application was developed to maintain a database of past drills, including metrics such as average player load, intensity, and duration. For each of the two training sessions prior to a match, the most relevant drill was identified based on the coach’s criteria, ensuring that training exercises closely aligned with match demands.</p><p>The findings of this study provide insights into how varying training loads impact match performance and offer a data-driven approach for coaches to refine training sessions, optimize player development, and make informed decisions regarding starting lineups and substitutions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

From Practice To Performance: Predicting Soccer Match Outcomes from Training Data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年3月28日

查看全文
  • Leili Javadpour,
  • Mehdi Khazaeli,
  • Ryanne Molenaar

摘要

This study analyzes performance data from training sessions of male soccer players to predict their match performance. Regression analysis was conducted to examine the relationship between training metrics—such as energy expenditure, sprint count, and power plays—and match performance features. Additionally, a binary classification model was employed to determine whether a player’s match performance could be predicted based on their training sessions leading up to the game. A correlation analysis using multivariate regression further explored the relationship between training session data and match outcomes.

To support training optimization, a drill selection application was developed to maintain a database of past drills, including metrics such as average player load, intensity, and duration. For each of the two training sessions prior to a match, the most relevant drill was identified based on the coach’s criteria, ensuring that training exercises closely aligned with match demands.

The findings of this study provide insights into how varying training loads impact match performance and offer a data-driven approach for coaches to refine training sessions, optimize player development, and make informed decisions regarding starting lineups and substitutions.