O’Sullivan, Hendry, Davies,...? A complex networks approach to ranking professional Snooker players

Abstract

The question is often posed regarding who is the greatest competitor within a given sport resulting in endless debates without definitive answer and is generally discussed via qualitative, rather than quantitative, arguments. In this talk we consider the problem in the sport of Snooker - a cue-based game with its origins in the late 19th century from the military bases of British officers based in India. A detailed analysis of matches played in the sport of Snooker during the period 1968-2020 is used to calculate a directed and weighted dominance network based upon the corresponding results. We proceed to consider a ranking procedure based upon a diffusive process that incorporates details of not only the number of wins a player has had over their career but also the quality of opponent faced in these wins. We demonstrate how this approach can be applied across a variety of temporal periods in each of which we may identify the strongest player in the corresponding era. This procedure is then compared with more classical ranking schemes. Furthermore, a visualization tool known as the rank-clock is introduced to the sport which allows for immediate analysis of the career trajectory of individual competitors. These results further demonstrate the use of network science in the quantification of success within the field of sport.

Joey O'Brien

Senior Quantitative Analyst

Senior Consultant in quantitative finance and data science for Acadia, working in R and Python.