It is often said that football is a simple game - and in some ways it is.
Yet it is also true that the world’s most popular sport is big business these days. In 2024 modern football incorporates elements of science, psychology, tactical theory and, increasingly, heaps and heaps of data.
In this article we will examine the field of soccer analytics and look at how data is revolutionising the sport. We will also look at the history of data in football and consider how the rise of artificial intelligence might impact the game in the years ahead.
The History of Data in Football
The earliest well-known case of data use in football came from Charles Reep. An accountant by trade who served in the British Air Force during the Second World War, he was also a huge football fan who was often frustrated with the way the game was played in England.
Reep preferred a direct, long-ball style of play and he produced analyses which purported to show the superiority of this style over a possession-based game. Working with Bernard Benjamin, Reep had a study published by the Journal of the Royal Statistical Society in 1968.
The duo’s work concluded that most goals were scored from moves containing three passes or fewer. Reep believed that the key to football was getting the ball into the ‘Position of Maximum Opportunity’ - close to goal in the opposition box - as often and as quickly as possible.
Reep’s work has since been questioned, but it was influential within certain parts of English football for decades. Indeed, the director of coaching for the Football Association, Charles Hughes, took on board many of Reep’s findings. Managers such as Dave Bassett and Graham Taylor, the latter of whom had an ill-fated spell in charge of England, subscribed to the thinking of Reep and Hughes.
Another early pioneer of data in football was Valeriy Lobanovskyi, a legendary manager who took charge of Dynamo Kyiv, the Soviet Union and his native Ukraine during a distinguished career that peaked in the 1980s.
In tandem with a scientist called Anatoly Zelentsov, Lobanovskyi used mathematical modelling to ensure his players were not being underworked or overworked in training. “Players have to react faster. Every moment, every action happens faster. Think faster, act faster – this is the football of the future,” he said.
AC Milan, the Italian giants, have also been ahead of their time on more than one occasion. First, in 1986, they set up the Mind Room with the help of Dr Bruno Demichelis, a leading psychologist.
Players had their blood pressure and breathing rates regularly monitored. Demichelis also interviewed Milan’s squad members to gauge their mood and feelings, then used his methods to quantify their answers. This data gave Milan an edge when they were the best team in Europe under Arrigo Sacchi in the late 1980s and early 1990s.
Later, in 2002, the club opened Milan Lab in conjunction with the Belgian doctor Jean-Pierre Meersseman. It was thanks to this creation that Milan’s players tended to extend their careers into their late 30s and early 40s - and data was at the heart of it. Because of the sheer amount of information Milan collected on their players, they were able to tailor fitness regimes and injury prevention methods to each individual.
As we can see from the above examples, data can be used in many different ways in football. The application of it has exploded over the last decade or so, and it is to the modern day that we turn next.
The Rise of Football Analytics
Every major club in Europe (and many minor ones too) now have data analysts working for them. Few stones are left unturned these days, and data is used in numerous different ways to try and give a team a competitive advantage.
There are also private companies operating in this area. One of the early market leaders was StatDNA, a US-based company founded in 2009. To demonstrate its influence, the organisation was bought by Arsenal just three years later.
“The company is an expert in the field of sports data performance analysis, which is a rapidly developing area and one that I, and others, believe will be critical to Arsenal’s competitive position,” said Ivan Gazidis, the club’s chief executive at the time. “The insights produced by the company are widely used across our football operations – in scouting and talent identification, in game preparation, in post-match analysis and in gaining tactical insights.”
The London-based Football Radar has not been bought outright by a club, but it too uses cutting edge-techniques to produce its findings. The company says it employs a “unique approach that combines subjective input from highly skilled analysts with innovative modelling techniques.” Football Radar employees meticulously watch matches and record their observations. The data is then crunched by statisticians to rate teams and players from around the world.
Statsbomb has partnerships with dozens of clubs around the world, including Roma, Celtic and Borussia Dortmund. Its website states that the company’s “data and IQ platforms allow soccer and football teams to get an edge on the competition in terms of player recruitment and scouting, performance analysis and opposition evaluation.”
It is not only professionals who have access to such information in today’s world. Platforms such as Wyscout, founded in Italy in 2004, provide match analysis, detailed statistics and hours upon hours of footage to anyone with a subscription. As a teenager, India-based Ashwin Raman landed himself a part-time role with Dundee United after impressing the Scottish side with the data-heavy analyses of players and teams he shared online.
Recruitment is an area of the sport where data is now seen as an essential tool. These days, the biggest clubs in Europe routinely sign players from all four corners of the globe. But this was not the case for much of football history. In the inaugural season of the Premier League in 1992/93, there were only 13 players from outside the UK and Ireland.
Back then, managers tended to be in charge of transfers. Along with their scouts, they would usually watch potential targets in person before making their judgement. But once the marketplace became global, this was no longer feasible. Coaches still usually watched videos of a player, but they are now just one voice in the recruitment process.
Sporting directors tend to lead the process these days, and most work closely with the club’s data departments. These employees have been known to unearth gems based on numerical information. For instance, if a team is looking for a tough-tackling holding midfielder, targets can be filtered down based on their performance across various relevant metrics. This process has the advantage of narrowing a field of thousands of footballers down to a handful.
Liverpool were at the forefront of the application of data in this manner. In 2012 they hired Dr Ian Graham as their director of research, shortly following the hire of Head of Analytics Michael Edwards in November 2011. He proved to be a huge part of the club’s success over the next decade, leading influential decisions such as the one which saw the Reds appoint Jurgen Klopp as their manager in 2015.
Graham and his team put data at the forefront of Liverpool’s recruitment. Famously, they persuaded Klopp to choose Mohamed Salah over Julian Brandt when the manager favoured the latter. The data suggested to Graham and his colleagues that Salah would be the better signing for Liverpool - and time has proven them emphatically right.
Data also helps teams analyse their opponents. In the past, scouting involved sending an employee to watch a team play once or twice in person. There is still a place for that, of course, but it is now combined with numerical studies detailing a team’s style of play, strengths and weaknesses. This provides coaches and players with a bigger picture.
What Does the Future Hold?
Football analytics has come such a long way in a relatively short space of time. Whereas the likes of Reep and Lobanovskyi did not see many of their contemporaries copy their methods, football clubs today are engaged in an arms race in the data field. With teams across the world determined to keep pace, new ideas and ways of working catch on very quickly.
It is interesting to ponder what the future holds for this field. Artificial intelligence continues to grow exponentially and it would be foolish to believe it will have no effect on football. Indeed, we are already seeing evidence of its impact. FIFA and UEFA use a semi-automated system to judge offside calls in their competitions. At club level, AI’s ability to process considerably more data than human beings is already being utilised by certain sides.
Take the example of Liverpool, who have been early adopters in this area too. The Reds have signed a deal with TacticAI, which uses Google’s DeepMind AI model to suggest set-piece routines. It analysed 7000 corner kicks taken in the Premier League to spot patterns and reveal which scenarios most often lead to goalscoring chances.
Two of the brains behind TacticAI, Zhe Wang and Petar Velickovic, describe it as such: “TacticAI is a full AI system that could give coaches instant, extensive, and accurate tactical insights – that are also practical on the field. With TacticAI, we have developed a capable AI assistant for football tactics and achieved a milestone in developing useful assistants in sports AI.”
Artificial intelligence can work in tandem with football analytics. As the data that is collected becomes more complicated, it will make sense to use AI to crunch the numbers. AI has a role to play in the collection of data too. For instance, clubs can use AI to provide more sophisticated tracking data to really break down their actions on the pitch.
There has also been talk of using AI to help the avoidance and management of injuries. “Sport by nature is unpredictable,” Dr Brian Moore, CEO of a bio-analytics company called Orreco, told the Guardian. “New AI tools can help the world’s greatest athletes reduce their risk of illness and injury, and play for longer.”
Of course, AI will pose its own challenges too. Data protection laws will need to be updated to ensure potentially sensitive and private information about a player’s health is adequately guarded. This is especially important when you consider that teenagers as young as 15 have been known to train with a club’s first-team squad. There are also issues specifically related to sport, which relies on certain values of fair play and, to an extent, a level playing field. If AI is misused, it could damage football instead of making it better.
No one knows what the future holds in sport and elsewhere. One thing is for sure, though: data will continue to revolutionise football in the years ahead.
