Multilevel Bayesian Modelling of Time-to-Serve in Professional Tennis

Tim Macuga Past Event

ACEMS & QUT Institute for Future Environments Guest Lecture

WHEN: Wednesday, 1 March, 2 pm - 3 pm

WHERE: QUT Gardens Point Campus, Room P419, Level 4, P Block


This presentation will provide an overview of the state of tennis analysis, and a summary of the major results and the opportunities for future work given the availability of point-by-point datasets for Grand Slam events. Tennis analytics needs improvement, judging by the current ranking systems for professional players. There is general interest in the value of pre-performance routines in elite sports and we use a Bayesian multilevel model to explore the time to serve of both ATP and WTA players in professional tennis. This study finds a wide variation in the pace of tennis players and how they adapt to game conditions such as the length of the previous point and the importance of the point.


Jim AlJim Albertbert is Professor of Statistics in the Department of Mathematics and Statistics at Bowling Green State University. He is the author of Bayesian Computation with R, and co-author of the books Curve Ball, Ordinal Regression Modeling, R by Example, and Analyzing Baseball with R. His interests include Bayesian modelling, statistics education, and the application of statistical thinking in sports. He has previously been editor of The American Statistician and the Journal of Quantitative Analysis of Sports. He is a Fellow of the American Statistical Association and recently received the Founders Award for distinguished service.