Modern football analysis is all about data science. Therefore, different models have been developed to describe and analyse the game properly. Last time, you got a little introduction into the xG model to understand a bit about the quality and use of created chances. To create chances, you first need to be in possession of the ball. One model to explain why a player can how and when get the ball, is the pitch control model, which will be introduced slightly now.
“Pitch control (at a particular location) is the probability that a player could control the ball, assuming the latter were at that particular location.” This is the definition by Liverpool FC’s leading data scientist, William Spearman. So the model describes the situation on the pitch in a particular moment. To get data out of the information pieces, you have to combine the different momentum pictures. Today’s standard creates 25 of such pictures per second, which leads to a data amount of approximately 2.5 million pictures per game that have are brought in order to create a match film. Those pictures are mainly filmed with cameras and transformed with AI into the pitch control model, which allowed coaches not only to use it in pre- or post-game-analysis, but also to use it for tactical changes during half time. The picture below shows, how such a momentum picture could look like. In this picture, the red team has little advantages in pitch control, but they are not big enough that they could generate obvious quick use of it.
Source: https://youtu.be/X9PrwPyolyU
Spoken easy, the player that has the ability to get fastest to the ball from his location is the player that controls the ball. The ability is determined by the player’s position, direction of movement and maximum possible speed. The target for a team has to be the control of most of the pitch or at least the most important regions of the pitch under its control.
As mentioned, not all points on the pitch do have the same value for a team. If team A is in possession of the ball, their own box is less valuable than any given point in their opponent’s half, because having the ball in team A’s box makes it a long way to score a goal. Therefore the “scoring opportunity” of every point on the pitch weights the value of the pitch control at various points. Such a model could for example be the xG model (for details check out our former blog article).
Pitch control does not necessarily lead to success. It is often related to ball possession, but there are numerous examples that neither pitch control nor ball possession automatically mean success. If a team for example play a well-formed counter attack strategy, they do not need much possession to be successful. No single metric can bring your team success. But the clear understand and wise combination of different metrics can higher your chance of being successful.
