Two Liverpool employees sat in the stands at Stamford Bridge at noon last Friday looking at the empty lawn pampered by six carts of LED heat lamps. One was Ian Graham, PhD in theoretical physics from the University of Cambridge. Another, William Spearman, a Harvard PhD from his work at CERN hunting for the god particle, the Higgs boson. François Englert and Peter Higgs predicted that the boson existed and won a Nobel in 2013. Spearman measured the mass and width of the particle. A few rows away, five stories above the grass of the Chelsea field, Leeds United sporting director Victor Orta was chatting with Guillermo Alonso, a collaborator.
The bleachers were populated by a rare mix at that hour: physicists, mathematicians, programmers, sports directors, scouts, assistant coaches. It was the lunch break of a day of debates on the points where the big data and football, organized by Statsbomb, one of the leading data collection and advanced metrics companies. Nobody compares time-of-possession percentages anymore. Goals don’t tell the whole story, but rather how much each shot in the game gets closer to the goal. And the kingdom of a thousand countries has ceased to be sought, because some countries are not the same as others. Not similar.
Javier Buldú, researcher at the Center for Biomedical Technology of the Polytechnic University of Madrid, presented a model that evaluates the decision of footballers when choosing a pass: compares the one they gave with those they could have given, in terms of risk and the value that add or subtract to a play. There was talk of measurements, location data, and also of the misunderstanding that still hangs between the scientist and the soccer player. “Translators are needed from one world to the other,” could be heard between chatter.
It’s not just about marginal gains – big improvements are possible
Ian Graham (Liverpool)
Vosse de Boode, the head of Ajax’s sports science and data analysis department, has been closing that gap for ten years: “I can knock on the coach’s door and say: ‘Look what we’ve found,’ she says. Like when they noticed something strange in how goalkeeper André Onana (Cameroon, 25 years old) faced close shots. He planted himself with his feet wider than was advised in the quarry. As they prepared to correct Onana, de Boode wondered, “What if he’s right?” They covered him with sensors, surrounded him with cameras and compared him to what the other goalkeepers did: Onana was the fastest, 20% faster. So, for eight weeks, what they did was teach the rest of the goalkeepers to do it like him. “I was surprised that the improvement was so great at this level,” says De Boode.
Ajax has traveled an unusual path in the use of big data. In general, the most powerful analysis departments, which in Spain hardly stand out at Barça, have been formed under the shelter of American owners, a culture in which data has been helping the business for years. As in the case of Liverpool, under the same corporate umbrella as the baseball Boston Red Sox, Fenway Sports Group. Ian Graham, the theoretical physicist who runs his research office, shared some clue: “It’s not just about marginal gains, but big improvements are possible,” he said. According to his calculations, raising the performance by 2% -3% can mean 60 million euros more in prizes for a club that plays the Champions League.
Operation Firmino
Although what he spent the most time on stage was recruiting: “Any analyst who is not working in recruitment is wasting time,” he said. His reports are credited with, for example, the final push to sign Jürgen Klopp, despite the fact that in his final year as Borussia Dortmund manager he only finished seventh in the Bundesliga. On stage, Graham recalled the signing of Firmino for 30 million euros: “A player from a team in the middle of the table in Germany, without experience in the Premier, who had not been international, who did not score more than ten goals a year … But he had things that were important to us and not so much for the market, such as the amount of chances he created and his strength in the air game, “he said. And in the corridors it was whispered: “It doesn’t even count 10% of what they do.”
Pablo Peña Rodríguez, innovation manager at Statsbomb, points out where he believes access to refined metrics leads the industry: “When enough competitors in a League adopt these new ways of proceeding, they are no longer optional, at least if you don’t want to compete at a disadvantage ”.
My dream is that the opinion does not have any weight in the decisions, that it is only the data
James Cryne (Barnsley)
The advantage that the orderly collection of data could bring was already sensed by Víctor Orta in the beginning under Monchi at Sevilla. By then, before the arrival of Statsbomb u Opt, the now Leeds sports director built an Excel with the scores given by the sports newspapers of the footballers in the games that he could not see. “That’s how we located Gameiro, because L’Equipe almost always gave him the highest,” he said. Now he is working on the development of his own mixed model for his club: “Why choose between data and people? We are not going to sign someone just with the data, and we are not going to sign anyone without looking at the data. There are things that do not appear in them, ”he said. Like the psychological analysis, fundamental when a player is transplanted from one place to another.
At his side on stage, one of the directors of Barnsley, Second, James Cryne is at an even more radical point: “My dream is that opinion does not have any weight in decisions, that it is only the data.” His conversion comes from a borderline experience. After watching for years as his family supported a club on the brink of bankruptcy, in 2017 they sold 80% to an investment group that included Billy Beane, the baseball executive who inspired Moneyball. Patrick Cryne, James’s father, indulged in the data to save the club, but James, despite his college math studies, was hesitant: “What if that’s not the way to do it?” He wondered. But now he’s convinced: if the team were to go down again, they have players to sell who retain value. While he remembers that in his moment of greatest crisis, only four were under 24 years old and had played more than 100 minutes: a wasteland more the result of panic than of analysis.
Also a big believer in data responses is Harry Moyal, deputy general manager of Olympique de Lyon, who signed him in 2015 from McKinsey. Before arriving at the soccer offices, Moyal was a conscientious group gambler, a member of a kind of football club. Now he can no longer gamble, so he set up a data analysis department in the club that helps the business and allows him to keep some of the fun: “Predicting the future is difficult, but not impossible,” he told the audience, before Explain your calculations on the longevity of the best returns. “A footballer who is among the 20% of the best has a 67% chance of remaining in that group three years later,” he said while explaining his method for calculating the financial risk of the club’s transfer operations, the game in the what bet now.
You can follow EL PAÍS DEPORTES in Facebook and Twitter, or sign up here to receive our weekly newsletter.