‘Big data’ – defined as large volumes of structured and unstructured data that can be amalgamated to provide new insights – has had a dramatic impact on the football industry. No matter what aspect of the game, from match prediction to player performance, ‘big data’ analysis has become an essential tool that can provide greater knowledge, competitive advantage and marginal gains to its various stakeholders.
With regard to match forecasting, some progress has been made in correlating the performance of football teams in major international competitions with factors such as home country advantage, talent pool and macroeconomic variables. PwC (Price Waterhouse Coopers), for example, have begun to publish annual reports that seek to predict match outcomes in international football. However, this is still more of an art than a science: PwC concluded in their 2010 Index that “it is difficult to produce reliable forecasts of the outcomes of the 2014 World Cup based on econometrics alone”.
A recent product of the ‘big data’ revolution in football is the concept of ‘expected goals’ (xG). xG is a metric used to determine whether a player should be reasonably expected to score a goal from a particular situation. It is calculated from analysing each shot a player makes and assigning it with an expected goal value (EGV). The next stage involves comparing EGVs across matches and making observations about the performance of the team and individual players. From the vast array of data collected, predictions can be made regarding the team’s performance in future competitions.
Development and implementation of the xG system is still in its infancy, and it therefore remains unknown whether it can be used to ‘beat the market’. Nevertheless, the potential is there for the raw data to be fed into algorithms to look for anomalies or irregular patterns. When controlling for stand out performance attributes, this technology could also be used to pick up on match-fixing and corruption in sport. Let us explain…
If the xG system is used in conjunction with other technologies such as video-assisted refereeing (VAR), irregularities can be uncovered in the performance of football players and referees. For instance, if a player suspected of match-fixing has an unusually low conversion rate for statistically ‘expected goals’, this could amount to a red flag that would warrant further investigation from the football authorities. It may even be feasible to use the data as evidence to support prosecution. Such a system could certainly help Russian football raise its professional standards and reputation in the run up to the Russian FIFA World Cup in 2018.
In 2016, PwC conducted an inaugural survey of major sport governing bodies around the globe, finding that the biggest concern was, above all, match-fixing. In recent years, corruption has been endemic in sport – not just in relation to match-fixing, but for major event bidding, elections and central resource distribution. PwC’s survey suggests that we could be reaching a tipping point in which public mistrust leads to sports fans turning away from major sporting events. The economic consequences of this would be disastrous for the industry.
Match-fixing has been a thorn in the side of football for years. In 2013, FIFA’s head of security described it as a “crisis”, while the head of the Asian Football Confederation claimed it is “pandemic” in the sport. In 2013-14 season, 13 matches were suspected to be fixed in the British leagues and it was little over a decade ago that the Calciopoli scandal rocked Italian football. Football players, referees and club officials are all vulnerable to seduction from corrupt organizations offering huge sums of money to alter their performance or actions. This is a global problem, however, that ‘big data’ and technology can help to solve. If used to identify irregular patterns and subsequently highlight corruption, the value of these technological developments will go above and beyond its ability to enhance player performance.
Still, the advancements that ‘big data’ and technology have made in enhancing player performance should not be understated. Innovative coaching teams around the world are harnessing the power of ‘big data’ analysis and technological resources in their obsessive pursuit to achieve marginal gains.
Julian Nagelsmann, for example, coach of Germany’s Hoffenheim, has been credited with revolutionizing training with the introduction of a 6×3 meter video screen on his training pitch. Connected to various cameras positioned around the pitch, the screen allows Nagelsmann to provide instant video feedback on where his players went wrong and how they can improve. Virtual reality headsets are also allowing players to simulate match situations, while technology is being used to monitor sleeping patterns.
Leicester City’s incredible march to the 2016 Premier League title was partly credited to data analytics. The club meticulously tracked data through a range of physical metrics that coach Claudio Ranieri used to tailor training programmes to the needs of individual players, thus reducing injuries. Indeed, Leicester suffered to the fewest injuries to their squad in the 2015-16 season, which undoubtedly played a crucial role in the club’s success.
‘Big data’ and technology has entered into the the world of football at an alarming rate, but it has the power to be a force for good. Although its ability to reliably predict matches remains limited, great progress has been made in enhancing the performance of football players and aiding referee decision-making. Further, combining the power of ‘big data’ through the xG system and VAR could help to stamp out corruption – an issue that has long plagued the beautiful game.
By Dr. Josh McLeod and Dr. Vince Hooper
Dr. Josh McLeod is a Lecturer in International Football Business at UCFB (Wembley). He has taught in a diverse range of subjects including business management, enterprise and business law. His core research interests are in the governance, finance and operations of football clubs.
Dr Vince Hooper is a dual British and Australian citizen and is a specialist in International Finance. He has taught at the top internationally ranked business schools in Australia, China and UK as well as publishing in top journals like the International Journal of Forecasting.