Fast bowling kinematics associated with ball release speed. An integrated data-science approach

Engineering of Sport 15 - Proceedings from the 15th International Conference on the Engineering of Sport (ISEA 2024)

Experimental research has widely investigated biomechanical variables associated with ball release speed (BRS) of fast bowlers. Outcomes of these investigations have identified that several variables have inconsistent associations with high BRS. Moreover, conventional statistical explanatory approaches such as correlation analyses and linear regression models generally employed to study these associations have demonstrated inherent predictive limitations. It is important to acknowledge that high explanatory power may not imply high predictive power. Applied data science techniques, specifically machine learning models, offer a promising avenue to gain a deeper understanding into these complex datasets. The aim of this research was twofold: 1) to identify movement characteristics that best predict BRS in cricket fast bowling and 2) to understand the relationship between these top predictors and BRS using an integrated data science approach.