Tactical analysis in football has undergone major developments in recent decades, largely powered by technology. In this context, video is the most effective way to communicate between coaches and players. Currently, the exchange of information can be enriched by graphs, arrows, heat maps and other types of displays of game statistics that are being used more and more frequently. However, video is still the ideal medium to convey concepts because it is the closest to the reality of football. In fact, from a very young age players develop an extraordinary spatial-temporal ability to understand videos.
At FC Barcelona, after each game, the coaching staff receive analyses of every game and every player. With the data generated from each game, the club’s data scientists have been able to work with the coaches to generate algorithms that help to understand the game of football in general and Barça’s DNA in particular. Gone are the stats that simply counted the number of passes, completed passes, or shots on target. With a simple option, the algorithm is used to obtain information on, for example, how long a player has been in a certain space or how many players he or she had the opportunity to link to, according to the criteria required by this system.
Where there used to be computers and video editors working for four or five hours to get this information from each game, today, all this can be automated and obtained in just five minutes. Algorithms developed by the club merely have to be applied to the recorded data and then you just press click. Using new data sources, such as positional data, the game can be contextualised and behavioural dynamics in space and time can be analysed.
Complex questions arise that imply a deeper understanding of the game: how do the players closest to the ball behave? Are they offering their team-mates passing options? What formations are used both with and without the ball? How do these formations change according to the opposition, the score, or the type of pressure received? How much does each player contribute to the team? Where is free space generated? And so many other aspects…
With these tools it is also possible to gain immediate access to the data for a whole season and link it with video to offer the coach the ability to interpret the game more broadly and completely. In general terms, at Barça the use of new technologies has been oriented towards the automation of technical work, which is achieved through the use of data in two dimensions. On the one hand, to understand the dynamics of the game from the intrinsic complexity of the sport; and on the other, to automate the descriptive analysis of the game that until now was done by humans who tagged moves.
This frees up the time for these professionals so they can spend more of their time doing jobs of greater value and that are more intellectually enriching. About three-quarters of a technician’s workday have been recuperated. Without spending nearly 80% of their time on mechanical tasks as they did before, they can now study, reflect, and analyse the game in much more depth. These are contributions to the scientific community that the club has publicly shared, and one of the premises of its research departments.
For its part, Metrica Sports has developed the Play software, an application that has sparked a paradigm shift at all levels of sport through technological democratisation, just like the club’s other programmes, including Pixellot’s AI-guided cameras that have been fitted at the Ciutat Esportiva Joan Gamper training ground. Just like Barça’s tool, it too uses algorithms to detect game events in videos, which means all the information that a coach might want to show his players on video almost is instantaneously available.
At present, Metrica Sports’ business strategy is focused on making these tools, that are being used in professional football, also available to countries with fewer resources and amateur athletes. “By empowering the grassroots, professional football benefits,” says Metrica CEO Rubén Saavedra. It is merely a case of taking advantage of the cheapening of processes brought about by technological evolution. Seven years ago, when Artificial Intelligence was taking its first steps, it might have cost five thousand euros to get tracking data from a match. Now it costs ten times less.
With this cost reduction, Saavedra and his partners came up with the idea of developing a technology to obtain tracking data from any video. When they posted their software on the Internet, they found that 95% of the downloads were by users who could not pay more than a thousand euros a year to get the data on their team’s matches. So they decided to go one step further and offer it directly for free. They only charge for options that can implement and increase the quality or accuracy of analysis, depending on requirements, but the basic functions are universally available.
This program has spread all over the world with more than ten thousand downloads. There are only two countries left to cover the entire planet. In places with fewer resources, the tool is growing exponentially as small football clubs have started to learn about it. A youth coach who combines his role with another job now finally has the same kind of time to analyse his matches as elite level coaches once had. Moreover, for young players it is just as essential as it is for professionals to be able to analyse their play with their coach using drawings and graphics, because this gamifies the study of sport, which is essential given the kind of stimuli that affect new generations.
There is also a return of knowledge. One of the most important advantages of putting a free program on the market is that it can learn from its users. To begin with, because a lot of people are using the Metrica Sports software for other sports, with examples on the net of it being used for cricket, ice hockey and even polo. This is a phenomenon that Metrica had already observed before by sharing anonymous match data for free and studying the new functionalities and ideas emerging from analysts who had the talent but not the access to data.
This tool has also been a revolution for scouting. Metrica has received requests from clients asking for videos of matches involving a player who has been producing interesting stats in order to use the software to obtain “context” data, i.e. how the player moves in relation to opponents, their maximum speed, physical load, metres travelled at high intensity per game, and so on. This is a significant amount of detail that was not so easy to detect accurately by eye alone.
A report on a player will always require a subjective assessment based on knowledge from human experience, but now scouts will be able to enrich their analysis with a much more complete and above all accurate view of a player’s characteristics. It also opens the door for retrospective studies. A club that has recorded all the matches played by a successful player that has come out of its academy can obtain all of that player’s data using this software and work out when he or she started to stand out. Youth selection and promotion processes will have fewer margins for error.
As technological advances allow for it, the ultimate goal of these tools is to provide data in real time, so this information can revolutionise the game. Improved technology will also permit greater accuracy in the identification of players following automatic detection, something that still requires certain manual corrections today. However, the most important medium term goal is the universalisation of the algorithm, i.e. to make software that analyses data according to predetermined criteria of a team’s style of play available to all, in such a way that a group of friends can have a kick-around in the park and then use a computer to extract the data and reach conclusions and indications based on the criteria of any famous coach they choose. The ideas of the most brilliant minds in the history of football will be at the service of any footballer who wants to improve their game.
Artificial Intelligence is far from being able to replace or even replicate human intelligence, but it can increase human potential. Key decisions can now be made much earlier. In more ambiguous aspects, there will be fewer doubts. Once these technological advances have been applied to football from the grassroots up, the quality of this sport will be boosted at all levels.
CATEGORY: MARKETING, COMMUNICATION AND MANAGEMENT
This model looks to the future with the requirements and demands of a new era of stadiums, directed toward improving and fulfilling the experiences of fans and spectators, remembering “feeling” and “passion” when designing their business model.
Through the use of computer vision we can identify some shortcomings in the body orientation of players in different game situations.
In the words of Johan Cruyff, “Players, in reality, have the ball for 3 minutes, on average. So, the most important thing is: what do you do during those 87 minutes when you do not have the ball? That is what determines whether you’re a good player or not.”
Muscle injuries account for more than 30% of all injuries in sports like soccer. Their significance is therefore enormous in terms of training sessions and lost game time.
DO YOU WANT TO KNOW MORE?
KEEP UP TO DATE WITH OUR NEWS
Do you have any questions about Barça Universitas?
- Research Center