Data Analytics in Sports: Football Applications
A Data Analytics department is quickly becoming a must rather than an innovation inside the football world. All the big clubs such as Liverpool, FC Barcelona, and Man City have invested heavily in this area in recent years, and even small clubs, like newly promoted Brentford FC, owe their success in big part to the use of Data Analytics. Last time, we talked about the applications of Data analytics in the business side of the organization; now, we are going to take a closer look at the possible applications of analytics in the sporting side of the organization.
Football was always believed to be too “dynamic” or “fluid” for it to adopt the use of Data Analytics. However, recent advancements in technology have proven that belief wrong. Tools like optical tracking are being used to track every player, every second of the game. Before, the only statistics that coaches and staff would receive were the traditional ones: shots on target, corners, ball possession, fouls, etc. Now, they can see every movement each player makes on and off the ball, the intention of each action, and even calculate the probability of each play ending in a goal. Additionally, analytics can be applied in areas like tracking a players’ performance and development, player recruitment, and injury prevention.
Game Tactics
As mentioned before, statistics such as shots on-target and passes completed have been in the sport for a while. The problem with these metrics is that they often fail to paint the full picture of the game. For example, if a team had 65% of ball possession, one might say just by looking at that number that they were in control of the game; when in reality, much of that possession could have been in their own half with unproductive passes between the defenders. Another example could be that a team had 10 shots, one would say that they had a lot of chances; when actually, most of those shots were from long-range, and they didn’t even trouble the keeper
With the new technologies, analysts have been able to obtain a much more detailed analysis of the game. They have developed metrics such as expected goals (xG) that have completely changed the way teams approach their games. The xG metric calculates the probability of a shot being converted into a goal, depending on the distance, angle, and other factors. It has been proven that long-range shots are quite ineffective and that the probabilities of scoring increase exponentially as you get closer to the box. This has made teams change the way they attack. They have shifted their focus towards a more “buildup” style of play, instead of trying their luck with unproductive long-range shots. Many other metrics like this one exist, and every team develops its own. It is quickly becoming a quest to see who has the better analytics team and who can extract the most relevant insights from the data. Manchester City has gone as far as hiring rocket scientists for their analytics team to get the upper hand over their rivals.
Something that the new technologies have also done is improve the way analysts communicate their findings in a way that coaches and their staff can understand. In the past, even if analysts found something interesting, they would often struggle to get the message across to the staff members. The use of video combined with heat maps, graphs, and other tools, has made it much easier to convey the information obtained. These insights can help teams understand how their rivals prefer to attack, how their formation changes in attack and defense, what type of runs each opposing player likes to make, and what positions the strikers are better at scoring from. All these things help teams to better prepare for games. For example, if you look at the Spanish La Liga, 10 years ago, the gap between FC Barcelona and Real Madrid compared with the teams at the bottom of the table was considerably greater than it is today. This is because even the smaller teams are using data to find ways to neutralize and exploit the weaknesses of their rivals.
Player Performance and Development
Similar to game tactics, data analytics can be used for player performance evaluations and player development. Through the use of video and data collected during matches and training sessions, coaches can help players understand what they are doing wrong or what areas of their games they should improve. With this data, coaches can also tailor training sessions to the needs of each player; for example, if the data shows that a team’s defense is struggling to defend when they are attacked from the wings, the coach can then design specific exercises to solve this problem.
Another way analytics can be helpful, is to address the performance of each player more accurately. Let’s say that a team is having a bad run of results and that their star midfielder hasn’t been contributing the number of assists he normally does. Using metrics such as the Vision Index (which tracks the quality and quantity of passes) or the Expected Assists ratio (which measures the chances of a pass turning into a goal), you find that the player has been performing like he normally does and contributing the same number of chances, only that the strikers are not taking advantage of them. This new information completely changes the scenario, and you can address the problem from a different angle.
Scouting and Player Recruitment
Perhaps the most important or most used application of data analytics comes when speaking of Player recruitment. Analytics has completely changed the way clubs scout for players these days. It is saving them time and money. With a large number of players in all the different leagues, it is almost impossible for a scouting team to track and watch the performances of all of them. Scouts can use analytics to filter through the databases and point out the players that meet the desired profile. Clubs can even create algorithms that will simulate how each player would adapt to the club’s playing system; as Raul Blanco from FC Barcelona explains, “before we sign a player, we must examine how he solves problems in the contexts he will face at Barcelona. It has become popular to categorize players using data without taking into account these contexts, but this distorts realities.” One thing that Raul points out, is the importance of interpreting data in the right way. Meaning, that there are a lot more things to consider other than goals, assists, and clean sheets when recruiting a player. For example, you are looking at a striker that scores a lot of goals, but he is slow, and your team has a fast counter-attacking style of play, then it doesn’t make much sense to recruit him even though his numbers are great.
Another way in which data analytics helps teams is with their financial planning. Nowadays, we see clubs pay crazy figures during the transfer windows, and it is easy for them to get carried away while trying to reinforce their squads. Analytics can help clubs to obtain a more accurate valuation of the players and, at the same time, take advantage of undervalued players. This can be applied to current and prospective players. For example, if a club owns a player and they receive an offer for him that according to the data is more than his actual value, the club can choose to sell him, and use the money to purchase another player who might be undervalued and that could fit in their system. This way, the club is getting a player that is compatible with their current system for a much lesser cost. Clearly, there are many other factors to consider when making these types of decisions, but data provides valuable information that can help with the decision-making process.
Injury Prevention and Workload Management
External Load tracking is becoming increasingly popular among clubs. Using GPS technology, teams can track the amount and the intensity of the work performed by their players on the pitch. This data can then be used to predict the risk of a player getting injured due to excessive workloads and plan accordingly to avoid it. Since each body type is different, this provides important insights for developing a tailored training schedule for each player to achieve peak performance.
Conclusion
Data analytics is changing the way football clubs operate daily. Throughout these two articles, we have seen the possible applications on and off the field, and we can see that there is data to be observed almost everywhere. Something that used to be an innovation a few years ago, today is a crucial part of every organization. Teams are investing more each time in this area to help them obtain a competitive advantage over other teams. This is an area that will continue to grow and one where specialists are going to be heavily required; we can already see specialists from other sciences, like mathematicians, astronomers, and computer engineers, migrating to this emerging field. And, as new technologies keep developing, one can only wonder where the limits to the uses of data are.
