Analyzing and predicting career trajectory of male elite junior tennis players: A machine learning approach
| Authors | |
|---|---|
| Year of publication | 2024 |
| Type | Appeared in Conference without Proceedings |
| MU Faculty or unit | |
| Citation | |
| Description | This study explores the intricate dynamics of the Junior-to-Senior (JTS) transition phase in elite tennis. Focusing on challenges faced by young talents, the research aims to unveil factors influencing successful transitions and the role of elite junior tournaments. In a retrospective-predictive analysis, 240 male tennis players from national teams in the World Junior Tennis Finals (2012-2016) were studied. The cleaned dataset (n = 2847) underwent statistical analyses, including Chi-square tests, Cramer’s V, Bayesian approaches, and Multinomial Logistic Regression (MLR). Artificial Intelligence (AI) models, using supervised learning classification, were applied. Results revealed 62.08% elite junior participants in the Association of Tennis Professionals (ATP) database, emphasizing the significance of team nominations and tournament results in predicting ATP status. Inferential and Bayesian statistics confirmed robustness, with MLR highlighting tournament results' importance. The most accurate AI model (2.1) achieved 84.5% testing accuracy and a 0.76 AUC, suggesting practical application. Findings underscore JTS complexities, emphasizing the pivotal roles of participation, national team nominations, and tournament results. The study recommends comprehensive player development programs, urging strategic team selections by national federations and academies. Coaches, stakeholders, and organizations should prioritize monitoring these variables for early talent identification and support. These measures collectively aim to optimize success trajectories, navigating the critical JTS phase in junior tennis players' sporting careers. |