Prestige Score: A Self-Consistent Method for Measuring the Performance of Tennis Players

• Activity Number: 485
• Type: Topic Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Sports
• Author(s): Filippo Radicchi*+
• Companies: Indiana University
• Keywords: Tennis ; Complex networks

Abstract

We are used to quantify the value of tennis players on the basis of absolute performances: total number of titles, total number of ATP points, . For example, we believe that Federer is the best player of all time just because he is the tennis player with the highest number of Grand Slam titles in the history of tennis. Such view, however, neglects the fundamental key of tennis: competition. The value of a player, in fact, cannot be considered as a quantity that does not depend on the quality of the opponents beaten during his career. Would Connors have won more Slams if he hadn't played at the same time as Borg, McEnroe and Lendl? Would Lendl have managed to win his 'nightmare' Wimbledon if he hadn't had to face the grass specialists of the moment Becker, Edberg and Cash? Our proposal for the measure of performance in tennis relies on the construction of a graph where players are nodes, and edges represents matches between players. The advantage of seeing tennis as a 'social network' is that the players are no longer considered as separate entities, but as part of the same social competitive system. In this way, it is possible to measure their performances not as absolute quantiti