Statistics in Sport

Date: Thursday 02 December 2021, 6.00PM
Location: Online
Online - joining instructions will be sent to those registered
Local Group Meeting


Share this event

RSS Edinburgh are excited to host three speakers who apply statistical methods to sport related research questions. 

Dr Guy Wilkinson – Lecturer in Sport, University of Stirling
Individual Player Contributions in European Football

Professor James Reade – Head of the Department of Economics, University of Reading
The Long Shadow of an Infection: COVID-19 and Performance at Work

Assistant Professor Johan Rewilak – Lecturer in Economics, Finance, and Entrepreneurship, Aston University

Performance at the Olympic Games: What we know, what we don't know and what we'd like to know.


 
 
RSS Edinburgh are excited to host three speakers who apply statistical methods to sport related research questions. 

Dr Guy Wilkinson – Lecturer in Sport, University of Stirling
Individual Player Contributions in European Football

 
This paper looks at applying new techniques to predict match outcomes in professional soccer. To achieve this models are used which measure the individual contributions of soccer players within their team. Using data from the top 25 European soccer leagues, the individual contribution of players is measured using high dimensional fixed effects models. Five years of results are used to produce player, team and manager estimates. A further year of results is used to check for predictive accuracy. Since this has useful applications in player scouting the paper will also look at how well the models rank players. The findings show an average prediction rate of 45% with all methods showing similar rankings for player productivity. While the model highlights the most productive players there is a bias towards players who produce and prevent goals directly. This results in more attackers and defenders ranking highly than midfield players. There is potential for these techniques to be used in the betting market as most models almost as well as betting firms.
 
Professor James Reade – Head of the Department of Economics, University of Reading
The Long Shadow of an Infection: COVID-19 and Performance at Work

 
The COVID-19 pandemic has caused economic shock waves across the globe. Much research addresses direct health implications of an infection, but to date little is known about how this shapes lasting economic effects. This paper estimates the workplace productivity effects of COVID-19 by studying performance of soccer players after an infection. We construct a dataset that encompasses all traceable infections in the elite leagues of Germany and Italy. Relying on a staggered difference-in-differences design, we identify negative short- and longer-run performance effects. Relative to their preinfection outcomes, infected players’ performance temporarily drops by more than 6%. Over half a year later, it is still around 5% lower. The negative effects appear to have notable spillovers on team performance. We argue that our results could have important implications for labor markets and public health in general. Countries and firms with more infections might face economic disadvantages that exceed the temporary pandemic shock due to potentially long-lasting reductions in productivity.
 
Link: https://www.dice.hhu.de/fileadmin/redaktion/Fakultaeten/Wirtschaftswissenschaftliche_Fakultaet/DICE/Discussion_Paper/368_Fischer_Reade_Schmal.pdf

Assistant Professor Johan Rewilak – Lecturer in Economics, Finance, and Entrepreneurship, Aston University
Performance at the Olympic Games: What we know, what we don't know and what we'd like to know.

During the Summer Olympics, euphoria surrounds forecasters who try to predict who will top the medal table. This excitement evaporates quickly, as the answers show little variation over time. Indeed, past success is the best predictor of future success. Statisticians that focus on between-country variation, also find different predictors of success than those who focus on within-country variation, as when quasi-demeaning the data, many previous drivers of success disappear. One phenomenon robustly associated with performing well at the Olympic Games is home advantage, which even improves a country’s attainment during the previous and following Olympiads. But there is still so much statistics doesn’t tell us, for example, what aspects of home advantage matter? In addition, different measures of success, provide statisticians different answers about which nations are the true Olympic Champions. I argue that whilst statistics in sports can be useful, there are times when statistics are unhelpful, and when examining statistics in sports, how simple is too simple. 



https://expertfile.com/experts/drjohan.rewilak/dr-johan-rewilak



 
 
Dr Guy Wilkinson – Lecturer in Sport, University of Stirling
Professor James Reade – Head of the Department of Economics, University of Reading
Assistant Professor Johan Rewilak – Lecturer in Economics, Finance, and Entrepreneurship, Aston University
 
Dr David Ewing for RSS Edinburgh Local Group