2023-10-13 04:23:27
Valverde arms his right leg and, with a subtle but powerful inside touch, shoots. The shot goes past a swarm of defenders, past the goalkeeper, hits the post and becomes Madrid’s first goal against Real. It is a great goal, a subjective concept, but one that any football fan can handle thanks to years of experience watching games.
According to the data offered by LaLiga in a replay, that shot had a 4.1% chance of ending in a goal. The statistic states that only 11% of the shots in the championship are scored, so anything below that figure is a shot with few options. Fabio Nevado, LaLiga analyst, establishes that 11% as the limit at which an opportunity begins to be clear; Cristian Coré, an independent expert who works as technical director of Driblab, raises it to 30%. That is to say, when a shot has a 30% or more chance of being a goal and it has not, the player has not been successful.
In reality, this sophistication, which can be difficult to understand, arises from a process similar to that of that fan who knows as soon as he sees it that the goal is a great goal. It is a question of memory or, as we talk about statistics, of database. An amateur has seen thousands of shots like that that went nowhere, intuitively his head comes to the conclusion that this shot has great difficulty.
Ferran’s shot to which LaLiga assigns a goal expectation of 7.6% ABC
The different statistics providers (LaLiga, Opta, Driblab…) do something similar but with more precision and a scientific basis. The difficulty of a shot is determined by a series of specific variables: how far the ball is from the goal, how many players there are between the ball and the goal, where the goalkeeper is or what angle the finisher has.
LaLiga, in collaboration with Microsoft, has a system of 25 cameras dedicated exclusively to accurately locating the players and the ball throughout the match. This tracking, known as ‘tracking’, allows us to know what the variables are in each shot. Other providers do not use such a sophisticated system, but the essence is the same. These data are compared with the shots that are recorded, hundreds of thousands, and thanks to that it is known that a shot like Valverde’s only ends in a goal 4.1% of the time. That is, for every hundred such shots, the player misses 96. That goal is, statistically, a rarity. Or, as they say in football language, a great goal.
Ana Rosa Victoria, Innovation Manager of LaLiga, explains why the championship offers a statistic like this: «We talked to the fans and they told us that they were interested, we tried to improve the graphics with their opinions. It is built with engineers, data and calculation scientists, that must be translated into a more understandable language. The objective of this metric was to convey the difficulty of scoring a goal, because we all believe that it is very easy, but in reality scoring a goal is very complicated.
There are doubts, even within the competition, that this cold mathematics has been able to penetrate the viewer. For many, the figure says nothing even though it is objective and provides data based on experience. “The problem is that there is a lack of context, we do some training with the narrators and a lot with the commentators, we give them data so they can give more clues about the teams,” explains Nevado.
LaLiga offers its service to clubs and broadcasters, but the general public can only access some selected figures that appear on a page called ‘Beyond Stats’, but there are other pages, such as Fbref, that use similar providers and allow access to statistics advanced. These data are common for most leagues, so if necessary they can be used to give clues to an unknown rival. Europa League or that newly signed Belgian league striker.
Many times all this only serves to confirm the obvious, that a specific goal was more or less simple. In football that doesn’t say much, but when all of a player’s shots are analyzed the conclusions are more robust. This aggregate statistics of shots is what is known as “expected goals” and is expressed with two increasingly common letters in football: xG. It is data that allows us to analyze whether the reason that player does not score is that he is failing a lot or if his chances are not good enough. Or whether a team is as good as it seems or is just getting lucky.
That is what the analysts, the technical bodies, look at most. «A player is supposed to score the same number of goals as the number of goals expected of him; If he is above we know that the player finishes well and if he is below we know that he is not”, explains Cristian Coré. To give an example, Rodrygo in the first nine days of the league scored a goal, but the aggregate probability of his shots (xG) is 3.4. That is, he should have scored at least two more goals. The figures are more correct the greater the amount of data and in the case of the Brazilian the suspicion is confirmed. Last season the sum of probabilities in LaLiga was 13.2, that is, it would have been normal for him to score 13 goals, but he only scored 9.
The Las Palmas mystery
This also applies to knowing the performance of the teams. Fabio Nevado explains it: «This year, for example, Las Palmas is having a hard time scoring goals. You can think if it is because he finishes little or because the quality of his finishing situations is low… all of that is what the metric measures. The Canarian team has scored six goals this season, but the quality of their shots says that if their players had been moderately accurate (that is, like a normal player, not like a star) they would have scored nine. This has a positive and a negative reading. The first is that, despite not scoring, the team creates danger. The bad thing, especially for a recently promoted player, is to realize that his attackers may not be good enough. And in the last two days they have been more successful.
The company Driblab, specialized in big data and analysis, has a repository that shows the players with the most expected goals in Europe so far this season. The list is led by Boniface (7,499 expected: seven scored), Haaland (5,739: eight scored), Guirassy (5,215: 13), Lewandowski (4,759: four) and Rodrygo (4.42 expected: only one scored), although, as As you can see, not everyone has scored as expected. Bellingham, for example, scores more goals than expected. He should have only scored four goals in the League, but he has achieved eight.
Expected goals based on the difficulty of their shots at this point in the season in European leagues
Because statistics help, but they do not offer anything substantially different from what experience offers. “When diagnosing a team’s completion problems, the xg helps you intuit before analyzing situations on video,” explains Nevado. Although he belongs to the LaLiga team that develops this tool, he is aware that it is an accompaniment, not a panacea: «We work with data every day, but for me the point has not yet reached where data is capable of providing a knowledge that a coach or analyst may not be able to obtain by watching the video. And yes, there are many things that you can detect by watching the video but it will never give you any information.
#enigmatic #fact #sentences #Rodrygo