Can data analytics improve a football club’s performance?
Initially making an appearance in the 2000s in America (particularly in baseball and American football), in which the stats are at the heart of the game, data analytics have come increasingly into play in the entire sports industry. Moneyball, the book from Michael Lewis, partly contributed to the data analytics democratization in the sector. Now, we can analyze data and statistics in almost all sports.
Soccer, one of the world’s oldest sports, entered in the data arena very late. The complexity of the sport and its rules, the importance of feelings and experience contributed to the slow adoption in this sector. However, the development of new technologies, innovations and obviously the work from different companies, such as Opta or Statsbomb, enable the use of data analysis in the industry.
When people talk about data in soccer, they often refer to the statistics heard the previous day on TV or seen in the newspapers. This data on ball possession, shots, the number of goals per player and club rankings and a plethora of other ratings, is relatively easy to understand. This data is called primary and secondary data.
However, there is also more advanced and technical data: the type of data generally used in clubs. It is often called tertiary and quaternary data. This kind of data is more complicated to understand and to visualize. It requires data collection tools and specific knowledge to analyze and to interpret.
In a club, the analysis of this data can improve match preparation and results. Indeed, to win a game, the team has to be well trained and well informed with as much information as possible about their future opponent. Coaches can now understand how their own players play, they have finally all the information they want about a player. This data analysis helped them to organise their line-up according to the skills of the players. They can also optimize their training and conditioning patterns.
PassNetwork and PassSonar are two of the advanced techniques of data visualization. The first one is more focused on the interactions between players whereas the second one shows the direction and effectiveness of players’ passes. These types of visualization are definitely useful to understand how the team play.
Secondly, data analytics is also useful in the recruitment of players. This aid to scouting is essential to maintain a certain competitiveness and to offer new options to the coach. Data analytics provides a new perspective and contributes to making better signings.
With these new tools and their own experience, recruiters can now find their rare germ. From companies who own the data, they can use the analysis software to analyze players they want, collect some information about a player target, be more accurate on a skill of a player. Thanks to these tools, they can target exactly the player they want by including filters on the research like “left foot”, “recovered balls” and many others relevant data.
The other way to monitor players’ statistics is to call a consultant. The 21st club is one of them. In order to satisfy clubs requests, the company built an analytics engine, named PIRLO (in memory of the Italian player), supplying data on 150 000 players. The tool calculates the link between players’ actions on the pitch and their team’s overall performance level, and assigns each player a rating. To suggest smart insights and recommend players, machine-learning algorithms are used.
Liverpool Football Club : A perfect example of the positive contribution of data analysis
In 2010, this club and its many trophies were sold to the American consortium New England Sports Ventures. The main shareholder of this organization, John Henry, brings a new vision to the club : Integrating data analysis into decision-making. A new director of research was recruited: Ian Graham. Graham rapidly became one of the key men in the club’s resurgence. The holder of a doctorate in theoretical physics from Cambridge, Graham is known to use data instead of video in order to analyze players.
With his own database, he helped the club identify undervalued talents who would fit into the lineup. Through his recommendations, the arrival of Naby Keita and Andy Robertson, two players unknown to the public, greatly contributed to the club over the last two years.
Graham was also behind the signing of Mo Salah. His data suggested acquiring this Egyptian player since he would pair well with Roberto Firmino, another player of Liverpool. Mohamed Salah finished first as the top scorer of Premier League (English championship).
Moreover, Graham has a role to play in sales decisions. Coutinho’s departure brought £170 million to the club — money which allowed them to recruit Alisson, Van Dijk and Fabinho. Three players in an unbelievable season last year : winners of the Champions League (European Cup) and second in the Premier League.
The use of data is clearly on the rise in football. Other applications prove the positive contribution of data in the industry. However, it is unlikely that data will take full control — it will never replace the experience and feelings of a coach, but it can play an important role in decision-making.
About this article
This article has been written by a student on the Grenoble Ecole de Management’s Advanced Masters in Digital Business Strategy. As part of a content creation assignment, students are given the task of writing articles based on their digital interests and disseminate the articles online. Articles are marked but we make minimal changes to the content. Thanks for reading!
James Barisic, Programme Director, MS DBS.
