West Ham vs Compact Midfields
Despite a strong run at the start of December, results and (more concerningly) performances have tailed off a bit over the last few games and it’s worth considering why that might be. Some of it is down to fatigue and to player availability but there’s one notable thing in common between the teams West Ham have struggled against: they all play with a narrow, compact midfield, predominantly within a 4–4–2 shape.
First, a bit of perspective — we’re 10 points better off than we were at this stage last season and Pellegrini’s points per game record (1.33) is already significantly better than both of Bilic’s last two seasons (1.18ppg in 2016/17 and 0.82 ppg at the start of 17/18) and David Moyes’ tenure (1.22 ppg in 27 games). This isn’t meant to be doom and gloom. It was always going to be a transition year as we bedded in half a new squad and our players adapted to a completely new style of play. That said, it doesn’t mean it’s worth writing off the issues we’re currently encountering.
Although it seems like an odd thing to say about a team that scored 3 goals in 3 successive games a month ago, a lot of our recent attacking play under Pellegrini has been lacklustre and much of that is down to the structure of our build-up play. In short, the issue is that it’s virtually all focused on progressing the ball down the wings.
If you watch us pass out from the back, you see a pattern emerge that repeats itself again and again: the centre-back moves the ball out to the full-back on his near side (sometimes by giving it to a CM who plays the wide pass), who carries the ball forward until they can combine the wide midfielder, who tucks inside slightly in order to combine with the FB down the flank. If we get stuck and can’t progress, the ball gets shifted to the nearest central midfielder who then shuttles the ball across to the opposite flank (often via his midfield partner) and we then try the same thing on the other side of the pitch in the hopes that the FB/wide midfielder can get us into the final third over there instead.
This passmap from the Palace game illustrates this structure quite nicely:
The lines represent 4+ passes between players and the thicker the line between two players, the more passes exchanged between the players in question. The thickest lines on the map are between the full-backs and wide midfielders on either side of the pitch, while the central midfielders have strong arrows connecting them to the wide players on their side, as well as the CBs, but very little else. It’s also noteworthy how isolated our strikers are — we struggle to access them when we’ve got possession and their lack of involvement is evident in the passmap.
Playing this way isn’t an inherently a bad thing. When we play against sides who commit a lot of bodies forward, focusing our passing out wide makes sense as it’s a good avenue for counterattacking. Likewise, when Felipe Anderson is having a good day getting the ball to him deep and early is useful as it lets him turn quickly and gives him more space to operate in compared to when he plays higher up the pitch.
The issue is that this style of build-up play is predictable, slightly one-dimensional, and easily negated by sides who willingly surrender space out wide in order to congest the middle of the pitch. I don’t think it’s a coincidence that our poorer performances lately have come against Palace, Burnley, Brighton, Birmingham and, to a lesser extent Watford. All of those sides play variants of 4–4–2 (although Brighton played with more of a 4–5–1 out of possession against us). The way they all organised their midfield into a narrow, compact unit forced us into a lot of sterile, horizontal possession and made it difficult for us to create good quality chances from open play. It might seem a bit unfair to include two games we won in this but if you look at the goals we scored against Palace (two long range strikes and a rebound from a free-kick) and Birmingham (corner and a back post header), I think you’d be hard-pressed to argue that we were especially good going forward in either of those matches.
The Brighton game was a bit different because they played with an extra man in central midfield, but it was the only game I could find a link to, so here’s a couple of examples that demonstrate what I’m talking about:
You can immediately see the shape that Brighton took up when we had possession in our defensive third of the pitch and how difficult they made it to play through the middle — all of the central midfielders are playing in close proximity to each other in a horizontal line that cuts off any passing lanes forward. Our attacking players in advance of the ball didn’t exactly help matters, as none of them are taking up good positions and offering our player in possession any options.
The two wide Brighton midfielders are sitting narrow to support the CMs, while being close enough to our wide players to press any pass out to the wing. The position Glenn Murray takes up here is really intelligent as well, as it stops Pedro Obiang from checking inside to spread play to the opposite flank and effectively funnels him into playing down the near side of the pitch. It’s something Brighton’s forwards did quite well throughout the game and when West Ham did manage to switch the ball to the other side of the defence, one of Brighton’s wider central midfielders (either Pröpper or Groß) would move forward out of the midfield line to do the Glenn Murray job of restricting the pass back inside, effectively changing their system to a 4–4–2 to compensate for Murray’s lack of mobility.
At the end of this sequence of play, Obiang shifted it sideways to Ogbonna to try to start our typical build-up pattern. There was nothing on though, so he clipped a long pass up towards Andy Carroll. Carroll lost an aerial duel to Shane Duffy and Brighton regained possession.
That pattern repeated itself consistently throughout the game, especially in the first half. 10 mins later we found ourselves in a similar situation with Ogbonna on the ball near the half-way line:
Although our midfielders managed to get this side of the Brighton midfield line this time, their opposite numbers were close enough to them to make a pass into feet risky and Murray’s position makes an inside pass difficult again. Solly March got caught on his heels from a deeper starting position so Cresswell initially gets free. Ogbonna passes the ball out to him but March is quick enough to close the gap to apply some pressure. The ball broke free to Anderson but March won a tackle and that allowed Brighton to recycle possession.
Contrast that with our shape out of possession when Brighton were building up from their goalkeeper:
Our two strikers are split and playing level with each other, so they’re not restricting access to the centre of the pitch; ideally you’d want them to be staggered so one of them can cover any Brighton midfielder that drops off. On the far side, Robert Snodgrass takes up a nothing half-and-half position — not close enough to press the Brighton full-back, not deep enough to cover his own. Felipe Anderson is doing the same thing on the near side but he’s out of picture. Lewis Dunk rolls a 5-yard pass forward into Dale Stephens who has an ocean of space to play in because no-one has tracked him or cut off a pass to him.
Stephens turns under no pressure whatsoever and by the time Obiang gets close to him the pass had already been played. Because Snodgrass fucked off up the pitch as part of that un-coordinated press, Locadia is free in behind him to receive a ball near the centre circle and Rice can’t cover enough ground to intercept. Locadia also has time to turn and dribble forward 30 yards unchallenged, before lifting a cross towards the far post. Brighton effectively shifted the ball from box to box through the centre of the pitch with three passes.
This approach to build-up play is partially down to the opponents we’ve been playing — because these teams are so disciplined and well-drilled and diligent in their defensive shape, we have no choice but to work the ball out wide to try and find a way around them rather than through them.
A bigger part of it though is down to our personnel, which makes me think that this pattern of play is a needs-must type of solution that Pellegrini has come up with. Being brutally honest, none of our players through the spine of the team are good enough passers to break down compact midfield units during spells of extended possession. Diop, Balbuena, and Ogbonna are all competent on the ball but have their limitations, whereas Rice, Obiang, and Noble all have their moments but don’t have the passing range to reliably play in a more expansive way through congested central areas. Noble came on and changed the game against Brighton and the two goals we scored came as a result of him bypassing their defensive lines, first by clipping the ball into a channel and then by quickly switching it out to the right wing before Brighton had time to set, but he just doesn’t provide that sort of thing regularly enough for us to play through the middle.
The centre forwards have been an issue recently as well. Their lack of involvement is down to our inability to progress the ball through the centre of the pitch, but it’s also caused by their own deficiencies. Carroll, Pérez, and Hernández are all terrible with their back to goal (although Hernández was showing signs of improvement in that regard), which means our central midfielders don’t have the option to bounce passes off of them to help us move through the phases of play that way. What makes this worse is that none of them offer anything defensively, so we essentially end up carrying two players who offer very little in or out of possession.
Having Arnautović back and fit will help as he likes to drop deeper and is adept at linking play. I always think that’s a bit of a waste though, as his best qualities are off the ball, using his physicality to disrupt opposition defences in order to create space for others or to create chances for himself.
Someone else who might help is Samir Nasri. On Saturday, we got a glimpse of what he could bring to side. Nominally playing as part of the front two, he essentially played as a third central midfielder when we were building out from the back to provide some connectivity to the rest of the side and, although it was against a lower standard of opposition, he was more effective in that role than Anderson and Snodgrass have been when we’ve tried to get them to do something similar this season.
How else could we solve it? The one time we look capable of progressing the ball through the middle of the pitch is when Issa Diop dribbles out of defence and the continues his run ahead of the ball. I’d like to see him given licence to do that more regularly, especially seeing as Rice is disciplined enough to fill in and cover for him when Issa goes on a mad one.
What we really need though is a central midfielder who is press-resistant and can regularly break the lines, either through their passing or through their ball-carrying ability. Players like that are expensive though and I can’t see us signing anyone who fits the bill in this window, so that has to be the priority for next summer. If Lanzini hasn’t been properly Dean Ashton-ed, he could be a game changer in that role. A centre forward with a good appreciation of space who can link the midfield + the attack and generally act as a foil for a more advanced striker would be nice as well.
The game against Arsenal this weekend will suit us as they play with a much more open combo in central midfield and the various systems they’ve been using have consistently left them frail at the back, especially in wide areas. We played well against them earlier in the season at the Emirates and Felipe Anderson in particular had a field day playing on the break.
Struggling to break down teams who play with a compact, narrow midfield isn’t a unique problem, but it is one West Ham are currently ill-equipped to deal with. We need to find a way to do that on a more regular basis if we want to be the possession-oriented attacking team that Pellegrini wants us to be.
