Football clubs, like any other company, constantly strive to find a competitive edge over their
rivals that will bring them success on the pitch and the extraction and utilization of data is fast becoming the key to dominance in the football industry.
Football data analytics involves the examination, converting and modelling of data that enables the finding of valuable information which enhances team performance, recruitment and injury prevention. Data in football is not a new phenomenon, companies first started collecting it in the mid-1990s. However it is only in recent years that clubs have fully begun to understand and appreciate it, recognising that it can greatly improve decision-making.
The Old Establishment vs The Data Nerds and The Impact of Moneyball
Football for so long was a game all about tradition. Unchanged for years, dominated by autocratic ex-players managers relying heavily on intuition with simple tactics like the long-ball game typically favoured. The application of data to the sport was dismissed due to the game being judged to be too fluid and chaotic, along with data scientists who were generally poor communicators being widely distrusted, particularly because they had never played football at the highest level.
Other sports, particularly those American, such as baseball and basketball, were streaks
ahead of football in terms of exploiting data analytics. The data revolution in baseball that
arguably sparked the global data revolution was initiated by Bill James. Through applying
the intellectual rigour and discipline common in academia with the analysis of data to baseball, he proved many widely held beliefs in the sport to be false. These baseball statistics became known as ‘sabermetrics’ and went on to have a profound impact on professional baseball.
Before the introduction of sabermetrics, teams relied on the skills of their subjective scouts to
discover and assess players. But after James’ publications, his followers such as Billy Beane,
who became general manager of Oakland Athletics, applied his analytical approach to scouting with great reward. Through using sabermetric principles to identify undervalued players, Beane was able to form a team that outperformed its wealthier rivals and became famous for winning 20 consecutive games in 2002. His story became the subject of the book by Michael Lewis, and later film of the same name ‘Moneyball’.
After the success of Oakland Athletics, the rest of the Major League Baseball teams followed in adopting sabermetric principles and harnessing the competitive edge of data analytics. Football teams at first continued to be suspicious of data and saw it more as a threat than a benefit. But over the years clubs have begun to adopt it and football has thus embarked on its own data revolution.
Where Is The Industry At Now?
Over the last few years the football data analytics industry has seen major development in the technologies enabling the collection, storage and analysis of data as well as a massive increase in the human capital devoted to it. This has resulted in vastly bigger and better datasets. When companies like Opta first started collecting match data, they gathered simple stats such as number of passes and tackles made per player. Now with tools such as optical tracking that registers the position of players and the ball 25 times a second, a few million data points are produced for each player every match. Other complex tools now measure defensive stability, pitch control and scoring opportunities off the ball.
Most top level clubs now employ their own data scientists to conduct analysis or even have an entire department dedicated to it. Clubs are using data in their everyday operations or even employing data scientists to carry out long-term research projects.
Omar Chaudhuri, Chief Intelligence Officer at Twenty First Group, acknowledged this in saying “clubs are using data scientists who are able to dig into a lot more detail and not just understand ‘The What?’ but ‘The Why?’ behind things and therefore can help inform tactical decisions, recruitment and so on in a much more granular detail.”
Omar Chaudhuri, Chief Intelligence Officer at Twenty First Group
Many clubs are also outsourcing their data analytics by working with specialist companies that provide data focused tools for opponent scouting and player recruitment. Smaller clubs can profit from data analysis too whereas once before it was limited to bigger teams. Thanks to cheaper and more accessible software being available that tends to be based on video footage.
Why Is Data Analytics So Important?
Data analytics has three major benefits for football clubs that enhanced performance analysis, improved scouting and injury prevention.
In-depth data analysis allows teams to evaluate past performances, addressing weaknesses and building strengths, as well as analysing opposition prior to matches.
During matches coaches are able to make tactical adjustments and substitutions through being fed data in real time from players using wearable devices.
Post-game, coaches can deliver analysis to the players usually with the aid of video footage and data, showing them where they excelled and where they faltered. Displaying these objective facts represents a very effective method of efficient and convincing coaching. All of this, most importantly allows, teams to form a winning strategy.
The implementation of data analysis in scouting and recruitment has enabled clubs to save huge amounts of time and money. Databases filled with detailed player profiles produced for clubs by in-house data teams or external companies allow them to filter through and pick out the suitable prospects, who are then evaluated further through scouts watching them. Rather than hiring numerous scouts to gather intel on hundreds of unknown players, clubs have access to this information quickly and cheaply. The beauty of having all the stats stored in a database means that scouts can analyse all the player’s previous matches rather than just basing a judgement from watching them once. More data gives a more accurate representation of the player leading to better recruitment decisions being made. Looking at the numbers also gives rise to unearthing undervalued players in the transfer market who go unnoticed under the human eye. Billy Beane was able to build his success at Oakland Athletics by exploiting this, and in football, teams such as Leicester and Brentford have profited greatly from it.
The sight of elite footballers wearing GPS vests under their shirts has become commonplace in recent years. These vests are used to measure players’ heart rate, distance covered and speed. This allows the monitoring and enhancement of athletic performance as well as the prediction of risk of injury. Observing players’ workloads enables coaches to predict their risk of injury, adjusting players’ training schedules and resting them where necessary.
Data Is Nothing Without Interpretation
Inspired by the work of Bill James in baseball, the last 25 years has seen the use of data become more and more influential in football. In the industry today, a staggering number of data points are produced for every match, analysed by a growing mass of data whizzes employed by clubs who now recognise the decisive effect data analytics can have on performance, recruitment and injury prevention. It must be said though that data without interpretation is meaningless. Fully grasping what the data actually indicates and converting the solutions it provides to manager and coach problems, back into understandable football jargon is not easy. Clubs are however getting much better at bridging the gap in communication between their football practitioners and data scientists.
Watch this space, as I am going to delve into the football data analytics scene further by examining the key trends, who the major players are and what the future holds in the industry.
Ben Arthur is an Unofficial Partner intern.
Comments