Monday can only mean one thing: noticing. And guess what? Today we’re noticing out of possession and HAHAHAHAAHA I LOVE OUT OF POSSESSION.

More specifically, #DidYouNotice how Manchester City have improved the defensive side of their game so let’s strap in and explore how. **

Let’s set a little bit of context. We all know how City have struggled with defensive transitions over the last few seasons…

…which is an issue a lot of elite teams who try to hold high territory possession have found: **

There are a number of explanations for why this is happening but the material point is that these high possession teams are struggling to deal with transitional moments. **

Last season, City could easily default to the “the squad is aged out” argument to explain their worrying rise in chance quality conceded through transitions but that has gone now they’ve gone through something of a refresh. **

At the beginning of the current season, it looked as though Guardiola was just falling straight back into his regular approach to out of possession (OOP) which is similar to the old in possession approach: try to control the opponent by keeping them in their own half. **

This means pressing high, often man-to-man, in a bid to prevent them from progressing the ball closer to the City goal. **

However, part of the reason why they were struggling last season is that Premier League sides are much happier going direct in these situations now, especially given the fact that these man-to-man tendencies in the modern game can be manipulated to the opposition’s advantage. **

To make it work, then, required increased intensity and commitment, which is something we saw from City in their early games this season (the games against Wolves and Brighton were notable examples). **

The problem was, City were still facing moments where opponents were playing through them, often with their back line isolated 1v1 against opposition attackers. **

Here’s the list of City’s Premier League games so far with the five worst performances per expected goals against highlighted. Notice how four of these games came in the first five games of the season. **

Clearly this aggressive approach early in the season was not working and so City have changed their approach somewhat. **

Since then, they’ve taken a more conservative approach, where they’ve tried to maintain a +1 advantage (an extra player) in their back line to offer a level of added defensive protection. **

But the problem with a +1 in the back line is that you have a -1 (a player less) somewhere else in the field which means, in principle, the opposition can generate overloads against City’s higher pressing unit. **

The question is: how do you prevent this from becoming an issue? How do you make it so the -1 higher up the pitch doesn’t just mean the opponent can just play through you easily? **

Well as we saw against Liverpool, the way they have been trying to do this is to keep the -1 weakness as far away from the ball as possible. **

To do this, they used a very narrow OOP shape (effectively a 442 shape with a midfield diamond): **

Of course, the narrowness of this shape looks like you’re simply ceding the wide areas, but the plan was precisely to lure Liverpool to move the ball into the wide areas, where City could then lock onto them in a man-to-man press. **

And whenever they lock on against the opponent in the wide channel, the City winger on the far side comes inside to make sure that there are no free players on the ball side. **

This does leave a free player for the opponent (the -1 that is the result of a +1 in the back line) but because they’re on the other side of the pitch, they are pretty inaccessible. So by allowing wide build up, they basically reduce the danger of the -1. **

This picture illustrates it nicely. Essentially, City are putting their -1 in the midfield line as a 3v4. But by moving the three across to the ball-side, they always leave the -1 on the far side away from the ball. **

Here’s what the full sequence looks like in practice:

**

It’s worth saying that this is a somewhat risky approach because it does mean you’re relying on your wingers (Doku mainly given City want the play on his side and Cherki) to maintain the correct position. **

If they don’t, the Liverpool FB could be accessible by a pass into the space beyond them which would put them into a 2v1 against the City FB. **

But City’s defensive structure is designed to mitigate this too. **

With Nico Gonzalez in the DM slot given a more zonal role, especially in deeper areas, he’s able to drop out in instances when the City full back has to jump up or when they are dragged wide, allowing them to maintain a +1 advantage in the back line. **

Here’s some examples of that happening:

**

Ultimately, this is a much more conservative OOP approach than we’ve come to expect from Guardiola. While there is still the opportunity for City to jump into a high m2m press, the structure also works nicely in a midblock shape. **

They can fall back into a more active form of midblock like this where Bernardo and Foden are ready to step up into the opposition’s first line of build up **

Or in winning game states if they want to take the sting out of the game, they can fall back deeper into more of a 4141/451 block **

In every phase of the game out of possession, they have different strategic solutions functional to their purpose. **

So what do we learn? Well, for all the chatter of Guardiola’s evolution on the ball this season, there is a parallel evolution taking place off it this season too. **

We’ve seen a number of games now where Guardiola is happy to allow the opponent to possess the ball without working hard to win it back immediately. **

In part, this is tied to the in possession approach e.g. playing more transitionally (artificially or otherwise) it’s advantageous to defend in a midder block to allow space to transition into. **

But most importantly, this represents (in my opinion) of an acceptance from Guardiola that he can’t control games 100% of the time in the modern form of the sport. Instead of trying to force the high pressing approach into an even more extreme iteration, this is a change of tack. **

• • •

Missing some Tweet in this thread? You can try to force a refresh