Creating Pass Maps in R

Pass maps are rather intuitive - they show where a player plays a pass from, and where to. Using Statsbomb's open data from GitHub, and following the excellent instruction from the BiscuitChaserFC blog, I am able to create a range of passing maps in R. These can be given further detail for completed and non-completed passes, and filtered down to specific events such as passes played by a certain player or when under pressure.

In this post, I have focused on Barcelona's 1-0 victory over Real Madrid at the Bernabeu, sending them 10 points clear at the top of La Liga in March 2019, as they went on to claim the title with the greatest historical points difference (19 points) between the two clubs to date.

Barca's 51% possession included 400 completed passes out of a total of 463. It's hardly surprising that this 51%, given the standard of opposition and playing away from home, falls a fair amount short of the average of 64.8% possession that Barca managed during the season.

Looking at the pass map like this, it is difficult to take much away from it. The greatest number of passes seem to be played just inside the Real Madrid half, suggesting that Barcelona were playing mostly on the front foot, and Real Madrid perhaps happy to sit a little deeper. We can filter the passes down to those played under pressure to get a better idea of this.

When playing under pressure, it shows that the majority of incomplete passes were long balls played forwards, and that as a team, they are possibly at a much lower risk of losing the ball by keeping to the short passing football that has characterised the side over the past two decades. Other teams may look at this pass map and see that quite often the unsuccessful long passes are played from the back, it is possible that putting pressure on the likes of Lenglet and Pique might be an opponent's best chance of forcing Barca into errors to give away possession, however, this is obviously a high risk tactic also, as the team is particularly adept at keeping the ball under pressure, and beating the high press could lead to overlaps and spaces throughout the pitch for Messi and co. to exploit.

---

Looking back at the first map of all of Barcelona's passes, it is clearly much busier down the left flank. Given that Messi was playing on the right of Barcelona's front three, some might assume that this must mean that Madrid did an excellent job at keeping the talisman out of the game, however when we look at his own pass map we can clearly see that this is not true.

The marauding number 10 was by no means limited to the right flank, operating over the entire front three area with free reign. This style of play has become common for Messi over recent years, as he acts as provider for Suarez, and adds on even more of his own, intent on carrying Barcelona to title after title. This free reign positioning causes difficulty for the opposition to know who's role it should be to pick him up, and his intelligence and brilliant movement limited Madrid to putting his passes under pressure on just seven occasions.

If we compare Messi's passing action areas to that of Dembele out on the left, we see quite a contrast, with Dembele sticking to his left flank, and his misplaced passes mostly look as if they were dangerous balls into the Real Madrid box. Dembele's passing map is what you would expect of a left winger, whereas Messi's is much more resembling of a true number 10, or attacking midfieder.

As it seems that Dembele's movement is mostly vertical, it is surprising that his opposite, Dani Carvajal, could not put him under more pressure on his passing actions than the freely moving Messi experienced. Both successfully completed six of their seven passes under pressure.

As Messi constantly comes off of the left flank, Reguilon, his opposite on paper, would constantly have been passing Messi onto a teammate inside in order to maintain the shape of the back four. This asymmetric attacking football is becoming much more common throughout top teams at the minute, as they look to overload certain areas of the pitch and create space to exploit elsewhere, so it is reasonable to think that Messi's movement is a ploy to pull Madrid's novice left back out of shape, in order to leave space for Sergi Roberto to push on forward. However, if we compare the maps of Roberto and Barca's left back Jordi Alba we see that there is no real difference in the balance between the two. If Roberto did push into the space left by Messi we would expect to see a greater number of his passes being played higher up the pitch. It could be suggested that the difference in space that the two fullbacks operate in is shown through their pass completion rates on the night; Jordi Alba had a pass completion rate of just 85.1%, compared to Roberto's 92.7%. Although both are still extremely high, assessing this over time might further show the potential joy that Roberto can enjoy down the right flank, with Messi giving him greater space to aid ball retention. At this point, it would be useful to go back and watch the match film to answer several questions that I have growing from this.

When Messi vacates the right wing, is there space for Sergi Roberto to better exploit? If so, how can Barcelona use this? And if not, where is the space? Was Roberto moving into the area that Messi left, but not getting the ball? Messi is undoubtedly the greatest scorer and creator in the history of football, so I would hate to be caught questioning what he is doing on a football pitch or suggesting that there was something he could do better.

This is further evidence of the need to always be able to join the numbers back to the context of the game itself to make insightful comments upon it. Although it has been great to use Statbomb's data repository to learn how to produce these passing maps, the questions and answers that we obtain are limited, and we can only scratch the surface whilst looking at such a randomly beautiful sport in this one-dimensional manner.

To further improve the passing maps, it would be useful to also label on pass success rates and make it clear that the direction of play is attacking the goal to the right on the image.