Over the last couple of blogposts we tried to evaluate offensive contribution. First on an individual level, then for the best duos in Europe.
Now, we will look at the most exciting part:
The offensive networks of the best European teams.
By now passing graphs are well established in football analytics. A great resource for these are for example 11tegen11 / Between The Posts.
These passings maps are based on all passes played during a game, regardless of whether they led to a chance. Because of this the central midfielder often happens to look like the most important player in a team. Or in this case the center backs, with Marquinhos being the “most central”. These maps do give great insight into how teams play, but we would like to take a bit of a different approach and focus more on who creates on offense.
That’s why we will instead use the created value metric, which will only include passes that led to a shot. Moreover, the passes will be weighted by how good of a chance was created and how difficult the pass was.
Players who are notorious for playing 5 meter passes sideways will not look good using this approach, even though they might have had many touches during a game.
The idea is, as you can probably guess from the title, to find out how teams create their chances and who are their go-to players for that.
Offensive networks of the champions
Over two seasons for the EPL, Bundesliga, La Liga and Serie A we could look at almost 160 offensive networks. When I flicked through all the different teams there are many peculiarities that would be interesting to look at, but I want to keep the posts as short and concise as possible.
Therefore we will here look first at the 4 champions of the past season and then at two more handpicked examples.
Please note, that to automate the process to produce these graphs, I do the following:
1.) Order all players according to their created value.
2.) Put the first player (with the highest created value) on the pitch.
3.) Only add the next player if he is further than 6 meters away from any player already on the pitch
4.) Continue until there are 11 players on the pitch.
This leads sometimes to goalkeepers being excluded and having 11 field players.
In general the offensive networks cannot be thought of as a line-up for a club, but are rather meant to visualize the importance of each player on offense as well as the interaction between all players in the team.
Juventus Turin: Cristiano taking over
At 33 in a new team, in a new league Cristiano Ronaldo immediately dominated on offense.
The size of the circle is proportional to a players’ created value on offense, the strokes between players show how often they connect with each other. Great duos would therefore have thicker lines to connect them.
Not surprisingly, Ronaldo is the center on offense. It’s even more interesting comparing it to the previous season, in which Juve (spurned on by the competition with Napoli) finished with 95 points.
This team did not have a clear superstar, but managed to score 16 goals more than in the following year. Generally they were more balanced on offense, with Dybala being more of a vocal point on offense.
Manchester City: The best offense in Europe
Pep good.
I was almost going to leave it at that. Man City just seems like an incredibly balanced team with multiple offensive threats. Very much different to how Juve’s offensive network looked last season. İlkay Gündoğan is not even showing up on this graph, because his position was too close to David Silva.
While Juventus offensive network seems to be focused (understandably) on Ronaldo, there is no clear vocal point at Man City.
Bayern Munich: From Heynckes to Kovac
Coming from Eintracht Frankfurt as somebody who had to prove himself at a big European club, while struggling with a new and improved Borussia Dortmund, Kovac had a lot of critics throughout the whole season. Even now after winning two titles, some questioned whether he would be allowed to return as head coach in the next season.
Looking at their offensive network this season, we see a team with multiple options on offense. Kimmich playing as an offensive fullback forming a great duo with Lewandowski, needs once again praise for his strong offensive season. Even though Bayern depends quite a lot on Lewandowski, they do have other options that can come over the wings. Pretty similar to what we’ve seen for Manchester City.
Since I did talk about Heynckes quite a bit, feel free to check how Bayern stacked up in the previous season. To be fair though, Heynckes was only there for the second half of the season, but still Kovac’s debut season doesn’t look too bad.
F.C. Barcelona: The expected
Barcelona is dominated by a duo, which we’ve already talked about in the last blogpost. There is a lot of talk about Suárez decline, however looking at the numbers we see that he is still imminent to Barcelona’s chance creation.
What does a bad offensive network look like?
Between the 4 European champions, to me Manchester City looks the most intimidating one.
Without seeing the offensive network of a properly bad team, we might underestimate the other European champions quite a bit.
Luckily, we can take a look at Hamburg’s 2017 team, which scored 29 goals over the whole season.
Kostic and Hunt being the only average players in the whole squad when it came to creating and finishing chances in that year…which wasn’t enough to keep them up in the league.
A quick glance at this chart puts Juventus’ offensive performance immediately in perspective, who had arguably the weakest offense out of the champions in the past season.
Other interesting teams
As said previously, I found some other interesting offensive networks while flicking through all graphs.
Two of them, I’d like to show here.
Udinese Calcio
I have to admit, I have not seen any Udinese Calcio game over the last two years (at least not that I can remember).
After seeing how Rodrigo de Paul was off the charts in terms of created value compared to his teammates I got interested. This graph is confirmation of how important de Paul was in the last season.
In a team where everybody else was at most average on offense, de Paul has an incredibly high (13.81 or 3.5x more than the average in his team) value for his team. This looks like the perfect example of a great player stuck on a bad team…but of course it could also be an above average player that shines in a team full of bad players.
Since I haven’t seen him play yet (except for some YouTube videos which could even make Valon Behrami seem like a great catch), I cannot judge. The numbers and thus the visualisation are impressive though, and it will be more than interesting to find out with which club he signs for over the summer.
Real Madrid: Post Cristiano
We’ve seen it from the other side for Juventus, but Ronaldo did also leave a significant impact on how Real Madrid has been playing this season. Partnered with the leaving (and later returning) of Zidane, their offensive structure changed quite a lot.
To be honest, even with Madrid struggling during the last season, I would have expected other players to pick up some of the slack.
Ronaldo leaving should have opened up space and opportunities for the rest of the team. This is not what we see. Scoring overall 31 fewer goals than in the previous season, the created value by all players dropped. Of course this cannot be all blamed on Ronaldo, since they went through some coaching trouble and many of their key players are aging, but Ronaldo having the ability to create chances for himself is definitely having an impact on some players looking worse this season than last.
If you are interested in any other teams, please let me know. I will share some more on twitter and I am also happy to respond to demands, since I already have the graphs available. Just shout at me and I’ll oblige!
Frustball 