Expected Goals: one of the reasons of the triumph of Atlético de Madrid

Publicado por LabHipermedia en

Atlético de Madrid ended up winning 2020/2021 LaLiga Santander title in one of the most equal and exciting seasons we remind. Sweeping at the beginning and suffering as only this team knows how to do at the end. There are many keys by which the men trained by Diego Pablo Simeone managed to prevail in the final sprint of the campaign against Real Madrid, Barcelona and Seville.

The goals of Luis Suárez, the saves by Jan Oblak, the tactical wealth of Cholo’s board or the growth of players such as Marcos Llorente or Thomas Lemar have been pointed out in many analyzes as determining factors so that the Colchoneros did not end up drowning on the shore after nine months of non-stop strokes, almost always ahead of his rivals. Although all this has been indispensable, it is worth asking: what can Big Data, of which we have recently talked in this article, tell us about Atlético de Madrid’s victory?

The rise of Big Data in football

Before delving into this question, let’s take a short parenthesis to point out the increasing visibility of Big Data in the game. What just five years ago sounded like science fiction is today an indispensable tool for club analysts. Big Data has meant that classics metrics (goals, assists, fouls, cards…) are no longer given so much importance, but instead of them, some parameters that measure in greater depth aspects such as high intensity efforts or creation data for a certain player or team are emerging.

And not only clubs or coaches make use of this type of information. As we said in “Big Data and football, an increasingly closer relationship”, in summer it was viral that Kevin de Bruyne got rid of agents to negotiate his renewal with Manchester City: the Belgian was sure that a pendrive loaded with gigs and gigs of information was his best chance to convince the Citizens that they should renew their contractual relationship. He may not have agents, but he does have a large team behind him that used advanced statistics as their main weapon to persuade his team.

Expected Goals, the fashionable metric

For this analysis, we are going to focus on the fashionable metric in the field of Big Data: Expected Goals or xG. Simplifying a bit, we could say that Expected Goals are a way of quantifying the value of a scoring chance, of answering the following question: what is the possibility that a certain shot become a goal?

To reach a specific percentage, Big Data uses a multitude of parameters to arrive at that value: distance and angle of the spiker with respect to the goal, part of the body with which the ball is hit, type of pass that precedes the throw, construction of the play (for example, if there have been rebounds, drives or dribbles prior to the shot) … A multitude of factors that are compared in databases with thousands of similar plays to determine the possibility that the goal go up to the scoreboard.

It is also necessary to point out that the Expected Goals statistic is applicable to players and clubs and, therefore, useful both for measuring individual and collective performance. It allows us to clearly see their effectiveness: a striker may convert many goals, but if his Expected Goals statistic is above that figure it may indicate that his precision in front of goal is not being ideal in relation to what he and his team are generating in attack. On the contrary, the fact that his Expected Goals are well below the totals it is indicative of the good moment of a scorer. These two examples can, of course, be applied to the overall team.

The most effective team

Having already defined what the Expected Goals are, it is now time to analyze the last LaLiga championship finished. We will use the data hosted on the ‘fbref’ website, which offers many Big Data parameters such as those already mentioned for free, and we will eliminate own-goal goals from the equation.

This is where we find one of the keys to Atlético de Madrid’s title: it is, without a doubt, the team with the most favorable difference between its expected goals (“only” 52.4, the fifth team in this statistic and far behind the 78.9 of the leader of this classification, Barcelona) and his total goals (65, second in this statistic only surpassed by the 80 of the team trained by Ronald Koeman). That is, he scored almost 13 more goals than expected based on the quality of his chances. Despite the fact that the clubs at the top of the table improved their records regarding this parameter, the Colchoneros were the team who made the most of it. The one that came closest to their effectiveness figures was Celta, although their 7.3 goals difference pales next to the figures of Diego Pablo Simeone’s team.

Two killers on the team

Of those 12.6 extra goals that Atlético scored to win the title, 5.4 are signed by Luis Suárez. The Uruguayan reached 21 goals, scoring the two that meant victories in the penultimate and the last day, showing that a striker of his class never loses his scoring instinct.

However, he is not the most accurate player in LaLiga. Not even from his team: Marcos Llorente takes those honors, since his 12 goals in the campaign far exceed the 4.3 Expected Goals of the former Real Madrid player. Not even Leo Messi, the precision made football player, comes close to the rojiblanco midfielder with his 6.4 goals difference between those made and those expected.

With all the aforementioned, it is less surprising that Atlético de Madrid came out champion despite having, on paper, fewer weapons than their two main rivals for the title. However, the precision of a surgeon with which he has moved throughout the year, making the most of his chances, has built the foundations of the victorious story of the Rojiblancos in the 2020-2021 season. Will Luis Suárez and Marcos Llorente be able to maintain that spectacular level of success in this year?

Categorías: Blog

0 commentarios

Deja un comentario

Tu dirección de correo electrónico no será publicada. Los campos obligatorios están marcados con *