In the era of big data, sports scientists are already using data in various areas such as load tracking, training programming, match simulation, and performance evaluation.

Technological advancements have expanded decision-making processes beyond coaches' knowledge and foresight. The belief that data science has no place in coaching is becoming less prevalent, although some coaches still hold onto this mindset. You may hear or read the phrase 'Get off the computer and watch the match!' from those who reject data-based decision-making, but it is important to note that modern sports scientists have adopted this principle.  It is crucial to maintain a clear and logical structure when discussing the benefits of data-based decision-making in coaching and to avoid biased or emotional language. The addition of further aspects must be avoided at all costs. However, it is common to encounter statements such as 'I do not require spreadsheets for coaching' or 'the coach's intuition is always superior.'

 

It is important to note that data and coaching should not only coexist, but also complement each other. With the advancement of technology, the amount of data that coaches and practitioners use to predict future events is increasing exponentially. In recent years, sports organizations, particularly in team sports, have begun to utilize sports technologies that can collect vast amounts of data, including health statistics and athlete movements that place a strain on the body during training and matches. The data is collected from individual athlete movements on the field and is used for various purposes, from training models to automatic tactical analysis. While some aspects of physical stress can be measured, such as distance traveled, high-intensity running, and sudden changes of direction, it is impossible to measure all physical stressors on an individual. However, by processing available big data with machine learning models, highly accurate predictions can be made about future performance and injuries. Machine learning models will bring revolutionary innovations to the application areas of sports sciences in the future. The increasing amount of data, parallel to the development of technology, will render classical statistics insufficient in processing the data. Therefore, it is safe to say that machine learning methods will be a necessity in sports sciences in the future.