Performance and characteristics of 1,602 tennis players

Data on 1,602 tennis players regarding their demographic and technical characteristics, performance and prize money earnings over the last 20 years.

The following dataset was created in the context of the topic “What statistics can tell us about the career of Stefanos Tsitsipas“: the purpose was to analyze the data by applying simple machine learning and statistical techniques in order to predict Stefanos Tsitsipas’ world rankings within the next decade.

Primary data was collected in March 2020 from the Ultimate Tennis Statistics website, which is licensed under a Creative Commons license (CC BY-NC-SA 4.0) and is based on open source software available on GitHub.

More specifically, we collected annual data for the period 2000 to March 2020 from the Association of Tennis Professionals (ATP) world rankings, as published on the Ultimate Tennis Statistics website. We then proceeded to collect profile data published on the same website, regarding 3,912 tennis players included in these rankings.

Indicatively, as can be inferred from Stefanos Tsitsipas’ stats profile on the website, each athlete’s “profile data” includes, among other things, information about the age of the player, the year he turned professional, the seasons he has played, his “backhand“, his favorite surface, the amount of prize money he has earned, the titles he has won, the best rank he has achieved, as well as his current rank, his ELO rank etc.

The original dataset, created as a combination of the above elements, could be described as a matrix with p columns and n rows – each row, one athlete, and each column one variable (characteristic). After computing the proportion of missing values across the dataset’s rows and the columns, we found that in several cases, missing values ​​corresponded to more than 50% of row and column contents. We therefore deleted said rows and columns, since the information they would provide would be negligible.

The result of the data cleaning process is the dataset listed below, where our analysis on the clustering of players was based.

Dataset columns explained

• names: player’s name
• prize_money: total prize money earnings ($)
• best_rank: best position held in the world rankings
• overall_surface_pct: overall success rate (all courts included)
• hard_surface_pct: success rate on hard courts
• clay_surface_pct: success rate on clay courts
• grass_surface_pct: success rate on grass courts
• ranksDiff1: first rapid ascent in the world rankings (difference in places held between first and second season)
• ranksDiff2: second rapid ascent in the world rankings (difference in places held between third and second season)
• ranksDiff3: third rapid ascent in the world rankings (difference in places held between fourth and third season)
• best_rank_std: best position in the world rankings adjusted (according to years of professional activity)
• age_turned_pro: age the player turned professional
• age: ηλικία του αθλητή
• plays¹: whether the player is left-handed or right-handed
• backhand²: one-handed vs two-handed backhand players
• favorite_surface³: player’s favorite court surface
• hard_titles_std: titles on hard courts, adjusted according to years of experience
• clay_titles_std: titles on clay courts, adjusted according to years of experience
• titles_std: titles the player has won in total, adjusted according to years of professional activity
• prize_money_std: prize money the player has received, adjusted according to years of experience

¹ Decoding: 1=Unknown, 2=Left-handed, 3=Right-Handed

² Decoding: 1=Unknown, 2=One-Handed, 3=Two-Handed

³ Decoding: 1=Unknown, 2=All-Rounder, 3=Carpet, 4=Clay, 5=Fast (H, G, Cp), 6=Fast (H, G), 7=Fastest (G, Cp), 8=Firm (H, Cp), 9=Grass, 10=Hard, 11=Non-Carpet, 12=Non-Grass, 13=Non-Hard, 14=None, 15=Slow (H, Cl), 16=Soft (Cl, G)

This is the result of a collaboration between iMEdD Lab and AUEB Sports Analytics Group research team aiming to promote robust quantitative analysis in sports, on both an academic and a professional level. The team works in the field of “Sports Analytics”, which includes the creation of statistical models and the production of predictions regarding sports results, draws on sports economics, and uses such tools as performance analysis, and visualization and measurement of competitive balance.




Translation: Anatoli Stavroulopoulou