Brand glossary of advanced analytics in football


Advanced analytics in 해외축구중계 in general and, specifically, in football, is becoming more and more present. At MARCA we have been gradually introducing new terms that are not yet completely familiar, but that we are sure will end up being incorporated into the colloquial jargon of the king of sports.

To continue to familiarize you with and better understand our analytical articles, we have prepared this glossary of terms, concepts and ideas to facilitate their understanding and which will be expanded as we introduce new concepts.

Expected goals

What are they?

Known in English as Expected Goals (and by its acronym xG ), it is the advanced metric par excellence in football. With it, it is a question of measuring the probability that a shot will become a goal based on a series of factors , such as the angle of the shot, the distance to the goal, the type of pass received, the part of the body with which it is finish off, if there were defenders in between, etc.

It is a mathematical model that analyzes thousands of auctions

With the same characteristics and thus compares it with the auction in question to offer a percentage of probability of success or, rather, a value between 0 and 1, where 0 is no chance of scoring. and 1 is a sure goal. For example, a penalty usually has an approximate xG of 0.76, that is, it ends in a goal 76% of the time.

We say that it is an approximate ago because each provider has its own model and, therefore, there may be differences in that probability, although they are usually similar.

It is important to note that the value does not consider the quality of the player’s shot. Evaluates only the action prior to the auction.

What are they worth?

Mainly to tell us the performance of a player or a team in front of goal based on the quality of the chances they have had. That is, it will give us a clearer idea of ​​how many goals he should have scored.

The xG of a certain occasion is not usually very representative because it can be influenced by luck. For this reason, a larger sample or a summation of xG of a player or a team throughout a match or, better yet, a season or tournament (the larger the sample, the better) is much more reliable.Benzema was not only the one who had scored the most goals (not counting penalties), but he had also scored 4.36 more than expected based on the quality of the chances he had. Vinci’s was the second who had scored the most, but, although he also performed well (+1.54), Correa and Lamella had converted in total occasions of greater difficulty. Meanwhile, Oyarzabal had four goals, only surpassed by the Madridistas, but due to the difficulty of those conversions he had 11 players ahead.

At the team level, it can be useful, for example, to know how a match has developed. The following graph shows us how Real Madrid-Elche’s chances developed throughout the clash. Each step is an occasion and the higher that step, the clearer that occasion was. That clarity is determined by the goal probability, that is, the xG.


This metric thus defined by the data provider Opts allows us to go one step further. There are other providers that call it differently but it is essentially the same, such as Expected Goals after the shot (Post Shot xG or PSxG), from the provider StastBomb.



Please enter your comment!
Please enter your name here