Back in June I attended the very successful Second Mathematics in Sport conference, organised by Ruud Koning on behalf of the Institute of Mathematics and its Applications and held in Groningen, Netherlands. Interesting papers were presented covering many sports. I particularly enjoyed the contributions in the areas of tennis, golf, soccer and cricket.
I attended a talk by Gordon Hunter, who presented results for modelling rallies using Markov chain models. He presented in the same session as Ian McHale of Salford University, who questioned how fair the UK golf handicap system is, and Sohail Akhtar, also of Salford, who considered the declaration strategy in cricket.
Benoit Jottreau presented a Brownian motion model for Betfair soccer prices before Dimitris Karlis and Ioannis Ntzoufras presented work on robust predicting of soccer matches. Their idea was to down-weight outliers, or freak results – generally soccer games with large margins of victory for one team. The models forecast very well apart from last season’s Champions League Final between Manchester United and Barcelona – the model down-weighted Barca’s clear superiority throughout the tournament which had produced many large victories. This shows the difficulty in forecast soccer outcomes!
David Percy also presented fascinating work on the handicap system for the Paralympic Games, highlighting the difficulty in knowing how much of a performance to attribute to a “good day” and how much to the inhibitions caused by disability.
These are just a few of the many innovative and interesting papers presented at the conference; the program is here.