Jan Wendt told Sky movement. "This is a huge leap in football data development." Wendt is co-founder of Plaier, a company that is helping to bring artificial intelligence technology to Premier League football.

They have worked with four Premier League clubs, which is 10 times the world. By collecting a lot of data and leveraging AI to understand it, they are coaching for transfers, the cohesion of the squad and now the dismissal of the coach.

Winter and his business partners initially planned to use AI as a prediction tool, but they quickly realized that merely measuring football is the main challenge for decision makers in the game. “It’s just as important to make better decisions now.”

They built what they claim to be the largest database of football ever - "exoticism of event data, injury data and salary data for over 100 leagues" - they are working with everyone from owners to data analysts to help change game.

"We look at it in three steps. First, we can tell you how strong your team is and how strong you need to achieve your goals. Two, we can tell you how much each player’s expected impact on team goals ends the season difference.

“We provide very clear insights so you know exactly where to aim to be stronger. The third part is helping the club find the right players and invest this money in the best way. A better insight is to help Making better decisions is basically better insights.

"Imagine you are a doctor and you have to diagnose without an X-ray. You know the leg injury, but you can only touch with your hands if it is broken or injured. Or someone has a lung problem, but you I can only listen to their cough.

"And then a guy with an X-ray appeared. It doesn't mean that the decisions from the past were wrong, it just gave you a lot more knowledge about what was going on. We provided the extra tool. It's for the owner Provides a very transparent view of what is happening in the club.

"If you want to get to Europe, these are your competitors, and that's how powerful they are. Want to beat them? How to get to the next level."

Although all clubs can access the same data through Plaier, they are used differently. Some are also willing to use it as a filter to better target their scouts, while others are more cautious and prefer the feeling checking as a signature there.

"They created their favorite player list, we created the recommended player list, and we saw if there was overlap. That's how the more cautious club operations are. The way to really use it is to really use it to find suggestions."

Surprise Kane prediction

Any lessons learned? "The defensive position is cheaper than the offensive position." Perhaps it's not surprising. However, the consequences of obtaining an expensive new striker can be complicated. In addition to goals, there are more them Fraction.

Some insights may be shocked. For example, when Harry Kane signed Bayern Munich, Plaier made headlines in Germany, when their model predicted that the Bundesliga record signature would not have a positive impact on the team.

"In the end, they had a 54 goal difference without him, and the goal with him was 49," Wendt said. Bayern had a minimum of 13 years. "Tottenham Hotspur also got better, winning more points, scoring more goals and having better goals differences. That's how we simulate it."

Kane is not very rare, just realizing that football is a team sport with many sports parts. Add something to the team and you may lose something elsewhere. "Martchester City also had fewer first seasons with Erling Haaland," added Winter.

Calculating individual player impact is very useful for clubs when making recruitment and retention decisions. "This is the most commonly used tool." Click the button and Wendt can display the results of replacing another player with another player.

"Event data have been around for 15 years or more. What we are trying to measure is the efficiency of the system, not how they dribble or how fast they are. How do players contribute to the success of their team? Exercise.

"It's important because it's a team sport with a lot of dependencies and correlations. What AI does is like a helicopter, and needs an overview. Then, it tries to find patterns in these billion correlations. Data, what important?

"If we propose a player to join Manchester City, we will perform 10,000 simulations of 38 games per Premier League team to understand how it works. You can't handle that data manually, it will kill a computer. Artificial intelligence is needed.”

He continued to speak his point with Manchester City. "For example, Ederson or Stefan Ortega? The score is almost the same." But then he moved Rodri from the city side. "It has a huge impact." This could have been marked as risk before he was injured.

"Players like Rodri are in the 7000 range on our metrics. There may be about 200 players in the world at the level of 6000, so there is a risk of a good player. It's very difficult to replace them. Can you replace them? Roger Federer is on the Davis Cup team?”

The coach's impact is overestimated

Contrary to Winter's view of the significant impact on world-class players, the coach's influence was overstated. He did not come up with the idea of ​​pre-thinking. This is a discovery. "It's really surprising for us," he admitted.

“One of the more interesting learnings gained from AI is that the quality of the player results in 90% of sports success, while the coach contributes only 10%. Now, remember that 10% of people can still make a difference.

"But coaches tend to be able to perform with the team for up to one year at most. In the past 10 years, there have been no coaches in our system that have performed well for a long time, with only a few exceptions." Who are they? “Christian Streich is in Freiburg and in Manchester City’s Pep Guardiola.”

Guardiola's recent struggles - of course without Rodri, can only emphasize that. It can work in another way, too. These models also show that contrary to the public opinion of the time, Jurgen Klopp did not perform much at Liverpool.

Wendt shows a graph that is at a distance in the quality team where the expected results are likely to be. “You can see Jurgen is here, just like last season.” The bank said the performance in line with expectations.

Again, this is not a discreditation of Klopp. Part of the reason these players are so highly praised is because Klopp has improved them. "Of course, Jurgen contributed to it. That's not to say Arne Slot is not worthy of praise either. But it's a very strong team."

“At the moment, I think our models show that Liverpool has the strongest team in Europe, perhaps the second strongest team. It might be surprising if the new coach comes in with new momentum, he can be able to add another 2%. ”

Now, the temptation to change coaches will grow, and the transfer window is now closed. "This is the only real thing you can change right now. It's like Hail Mary in a football game." But with the insights provided by Plaier, hopefully, better decisions will be made.

"The owner always has this problem, is my team not good enough, or is the coach not getting the best performance from the team? Ever was."

Winter talks about the “education market” and “providing a more precise decision-making basis for decision-making”, which can still be guided emotionally and intuitively. Some suspect this may be just the beginning, as Plaier intends to expand his influence.

"We have some psychological parameters right now. We can predict harm. It's an extraordinary tool, we add things every few weeks, but be careful because there's too much. It's hard to operate."

It seems that football has become more complicated. Ironically, all of this is to make decisions simpler. “We want to do black and white for the club. This player is either good enough or makes you better. Welcome to the world of AI.