Scoring probability model based on service landing location and ranking points in men’s professional tennis matches

Abstract

Service is regarded as the most crucial shot in tennis matches, and hence affects the outcome. This study aims to construct a mathematical model that expresses the scoring probability of service shots. Here, “scoring” means that the server wins points in subsequent rallies after a service. The explanatory variables are the service landing location and the ranking points of the players (logarithms of ranking points for the two players). The model construction method is a stepwise regression of the generalized linear model. The model is constructed using more than 110,000 service shots from Association of Tennis Professionals Tour tournaments of 2019. The model comprises four service situations: first/second and ad/deuce sides. The developed model reveals the following facts: in the first service, advantageous locations for the servers are not symmetric between the advantage and deuce sides. Higher-ranked players can win more points than lower-ranked players even if their shots land in easy locations, that is, in the middle of the service court.

Keywords

Ball tracking system racket sport sport analytics

Get full access to this article

View all access options for this article.

References

Johnson

McHugh

Wood

et al. Performance demands of professional male tennis players. Br J Sports Med 2006; 40: 696–699.

Akinohi. https://logmi.jp/tech/articles/324033 (2021, written in Japanese. Accessed 14 Jan 2022).

Whiteside

Reid

. Spatial characteristics of professional tennis serves with implications for serving aces: a machine learning approach. J Sport Sci 2017; 35: 648–654.

Born

Malejka

Behrens

et al. Stroke placement in women’s professional tennis: What’s after the serve? Int J Racket Sports Sci 2021; 3: 37–45.

Hawk-Eye. https://www.hawkeyeinnovations.com/sports/tennis, archived in https://web.archive.org/web/20210815194835/https://www.hawkeyeinnovations.com/sports/tennis (accessed 31 Oct 2021).

Mecheri

Rioult

Mantel

et al. The serve impact in tennis: first large-scale study of big hawk-eye data. Stat Anal Data Min: ASA Data Sci J 2016; 9: 310–325.

Mather

. Perceptual uncertainty and line-call challenges in professional tennis. Proc R Soc B: Biol Sci 2008; 275: 1645–1651.

Sackmann

. JeffSackmann. GitHub. https://github.com/JeffSackmann (accessed 28 Oct 2021).

ATP. Rankings — FAQ — ATP Tour — Tennis. https://www.atptour.com/en/rankings/rankings-faq (accessed 15 Jan 2022).

10.

Konaka

. Match results prediction ability of official ATP singles ranking. Preprint on arXiv, https://arxiv.org/abs/1705.05831 (2017).

11.

Kovalchik

. Searching for the GOAT of tennis win prediction. J Quant Anal Sports 2016; 12: 127–138.

12.

ATP. Results Archive — ATP Tour — Tennis. https://www.atptour.com/en/scores/results-archive?year=2019 (accessed 28 Oct 2021).

13.

Tennis Abstract. ATP Lefthander Rankings. http://tennisabstract.com/reports/leftyRankings.html (accessed 21 Jan 2022).

14.

Mathworks. stepwiseglm. Create generalized linear regression model by stepwise regression. https://jp.mathworks.com/help/stats/stepwiseglm.html#d123e768995 (accessed 23 Mar 2023).