A detailed analysis of game statistics of professional tennis players: An inferential and machine learning approach

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14510%2F24%3A00137549" target="_blank" >RIV/00216224:14510/24:00137549 - isvavai.cz</a>

Výsledek na webu

<a href="https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0309085" target="_blank" >https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0309085</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1371/journal.pone.0309085" target="_blank" >10.1371/journal.pone.0309085</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A detailed analysis of game statistics of professional tennis players: An inferential and machine learning approach

Popis výsledku v původním jazyce

Tennis, a widely enjoyed sport, motivates athletes and coaches to optimize training for competitive success. This retrospective predictive study examines anthropometric features and statistics of 1990 tennis players in the 2022 season, using 20,040 data points retrospectively obtained from the ATP official source after the end of the season. These data were cross-verified with information from other sources before categorisation to address any discrepancies. Employing various analytical methods, the results emphasize the strategic importance of tournament participation and gameplay for financial gains and higher rankings. Prize money analysis reveals a significant disparity favoring top players. Multivariate Analysis of Variance highlights the need to consider multiple variables for understanding ATP rankings. Multinomial Logistic Regression identifies age, height, and specific service-related metrics as key determinants, with older and taller players more likely to secure top positions. Neural Network models exhibit potential in predicting ATP Rank outcomes, particularly for ATP Rank (500). Our results argue for the use of Artificial Intelligence (AI), specifically Neural Networks, in handling complex interactions and emphasize that AI is a supportive tool in decision-making, requiring careful consideration by experienced individuals. In summary, this study enhances our understanding of ATP ranking factors, providing actionable insights for coaches, players, and stakeholders in the tennis community.

Název v anglickém jazyce

A detailed analysis of game statistics of professional tennis players: An inferential and machine learning approach

Popis výsledku anglicky

Tennis, a widely enjoyed sport, motivates athletes and coaches to optimize training for competitive success. This retrospective predictive study examines anthropometric features and statistics of 1990 tennis players in the 2022 season, using 20,040 data points retrospectively obtained from the ATP official source after the end of the season. These data were cross-verified with information from other sources before categorisation to address any discrepancies. Employing various analytical methods, the results emphasize the strategic importance of tournament participation and gameplay for financial gains and higher rankings. Prize money analysis reveals a significant disparity favoring top players. Multivariate Analysis of Variance highlights the need to consider multiple variables for understanding ATP rankings. Multinomial Logistic Regression identifies age, height, and specific service-related metrics as key determinants, with older and taller players more likely to secure top positions. Neural Network models exhibit potential in predicting ATP Rank outcomes, particularly for ATP Rank (500). Our results argue for the use of Artificial Intelligence (AI), specifically Neural Networks, in handling complex interactions and emphasize that AI is a supportive tool in decision-making, requiring careful consideration by experienced individuals. In summary, this study enhances our understanding of ATP ranking factors, providing actionable insights for coaches, players, and stakeholders in the tennis community.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

30306 - Sport and fitness sciences

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Plos One

ISSN

1932-6203

e-ISSN

1932-6203

Svazek periodika

19

Číslo periodika v rámci svazku

11

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

27

Strana od-do

—

Kód UT WoS článku

001349526900029

EID výsledku v databázi Scopus

2-s2.0-85208360801