Games between top six sides do not tend to look like this:
That's just the first half, but the whole of Liverpool against Arsenal very much looked the same. A remarkable congruence of tactics and the new goal kick rule that meant that Arsenal played really, really deep and essentially invited Liverpool to play ahead of their box, which in the main they did. Seriously, even Jose Mourinho back in late 2017 parking the bus at Anfield while shopping for a 0-0 and getting one looked very different, this Arsenal set-up was something else entirely:
Was it fear? Last December, Liverpool hosted Arsenal and Unai Emery was chastened by a 5-1 drubbing that was 4-1 at the half with a pass network that looked like this:
Just too open? Interestingly, only five players started both fixtures for Arsenal, which feels representative of the player flux that we've seen at the club.
What really piqued my interest about last weekend's game was the half time shot count. Arsenal conceded 15 shots while taking five of their own in a half that as we've seen was played predominantly in one part of the pitch. Now the expected goal values of all these shots were closer together, Liverpool's 15 shots rated a mere 0.83 while Arsenal's measured at 0.6. So through one lens--expected goals, this was a close half. By other lenses, it wasn't. Many, many first halves of football finish with the two sides within a quarter of a goal in xG, just as in this game. Last season there were 122 in the Premier League alone. Of those 122 games, the maximum shot differential at the half was eight. So already here we have an outlier game: very close xG but not remotely close shot counts are scarce. How often did a team ship 15 shots in a first half last season? It's well known that shot rates tend to increase as matches progress, so even when you see a 30-plus shot beatdown, most of the time, the shots get racked up in the second half. And that's true here. Only four times in 2018-19 did a team reach the half time point having allowed 15 or more shots. Three of those matches were small teams flailing around against Manchester City and the other, oddly, was Newcastle versus Huddersfield. All this brings to the boil a personal frustration of mine. Expected goals is a useful tool in many aspects of analysis, but it's not the only one. What actually happened in this half of football over and above the modelled value of the chances?
To allow 15 shots of any kind in a half is a clear sign that a defensive plan isn't working. Take Liverpool's left flank, expertly patrolled by Andrew Robertson and Sadio Mané with Georginio Wijnaldum shuttling across. It was very easy for these guys to progress, here are Robertson's first half ball carries and passes:
Long carries down the flank, deep into Arsenal territory, which on another day may have caused greater pain. Four times in that half, Robertson put balls into the box that went straight through. Twice, David Luiz judged to let the ball through with players lurking behind him. Mané misplaced two open play passes in the half, Wijnaldum just one. This flank was just open and easy to travel through.
If we evaluate the half we can see a theme of space and time.
Here are my notes:
2nd min: Robertson advances, fizzes the ball across goal, Firmino behind Luiz stretches but can't get there
3rd min: Set piece, through the box, Luiz lets it go again
8th min: Alexander-Arnold open cross to nothing
nb. Three times in opening 8 minutes Matip is making passes on edge of final third, Arsenal super deep
11th min: Adrian error, Aubameyang over
13th min: Mané gets in round the back to nothing
15th min: Robertson advances, 3 seconds nobody comes to him, cross/shot
16th min: Robertson to an open Mané on halfway line, Mané carries to edge of box unchallenged
20th min: Robertson cross, straight through
21st min: suicidal clearance, good chance for Mané, saved
22nd min: Salah shot from 30 yards, no pressure on him, blocked
22nd min: Robertson low cross, straight through
23rd min: Firmino, skill, space in box, shot blocked
24th min: Mané open but wide, skies it
nb. Arsenal when trying to attack spread right across the pitch, no linkage
30th min: Pépé bends one from range
33rd min: Henderson error, Pépé misses highest value chance of the half
At this point it's seven shots to five, and perhaps Arsenal feel good, but time and again, Liverpool have found space in good positions, they just haven't done anything really useful with it or contrived to convert. Then the last ten minutes of the half are a complete disaster for Arsenal.
In the 36th minute, Liverpool cross the half-way line in possession. A shade under five minutes later, they score.
In the interim, Arsenal do not touch the ball outside their own half. These are their open play passes during that phase:
That's four completed passes and two failed in five minutes of play.
Liverpool found a way to enact extended pressure and finally were rewarded for it. Every time play restarted, they were faster to the ball, and quicker to react, as you can see in the notes:
36th min: Liverpool complete 5/6 passes, Alexander-Arnold overhit cross.
37th min: Arsenal win consecutive throws then give away ball with aimless punt to halfway, Liverpool complete 4/5 passes, keeper claims Robertson cross, Liverpool quickly win cheap turnover and are fouled on edge of final third, centrally.
38th min: Robertson high free kick cleared by Luiz, Liverpool win corner, Van Dijk header, blocked, Alexander-Arnold speculative volley, blocked out for a throw.
39th min: Salah snap-shot from throw in.
40th min: Arsenal try to play out, fail, Liverpool win ball back comfortably, Firmino makes space outside box, shoots, blocked, recycle possession, Alexander-Arnold picks up ball, crosses, blocked out for a corner.
41st min: Alexander-Arnold corner, met by Matip (Van Dijk being fouled), goal, 1-0
At this point, Liverpool have added five more shots, and they were to add three more in the remaining minutes of the half, while Luiz also cut out a Firmino throughball and Salah found the return pass from a one-two with Mané nipped off his toe. Without ever having to really exert themselves extensively, Liverpool turned the screw and finished the half extremely strongly. The next shot they would take would be the penalty to make it 2-0 and effectively seal the game.
We can see Liverpool pressed high up the pitch during that first half--they could hardly do anything else--and Arsenal very much allowed them to:
They let Liverpool push onto their full-backs and this made outlet balls difficult to complete.
The point here is that the outcome of this half was inevitable when factored against the tactics Arsenal deployed. It's easy to grasp the Pépé chance and feel it was potentially decisive, but it's only a one in three opportunity, and he's not even in the box. A style of play that struggles to even gain possession outside its own half is fairly doomed to failure. We saw the same the week before during Manchester City versus Tottenham. Here's Tottenham's pass map for minutes 30 to 50:
Does it matter? Well yes, Tottenham allowed nine shots with an expected goal value of 0.83 during this period, and Sergio Agüero scored. Nobody is suggesting that matches away at Manchester City or Liverpool are easy to approach but these Tottenham and Arsenal performances certainly land on the leaner side both stylistically and via fundamental metrics. By attempting to outflank the best counter-pressing teams in the league with neat play around your own box, you're almost certainly inviting trouble. And I think this is what frustrates me in particular about Arsenal's approach. Liverpool had shown a potential defensive weakness in their first two games in the Premier League this season. Norwich scored and hit six shots on target from 13 with 1.1 xG, while Southampton also scored and landed 1.6 xG from their 11 shots. Unai Emery's gameplan felt like one which respected the Liverpool team of last season, the team that was firing on all cylinders when demolishing his team 5-1, not the team that hadn't quite clicked into gear in 2019-20. They afforded Liverpool too much respect while negating their own strength--their attacking prowess--and ended up on the wrong end of a one sided match regardless.
Emery has been accused of being timid before, especially in away games. It's an obvious area to work on for the future. These points may well have been dropped regardless, but in the long term, any ambition towards progress and success will dictate that his team is more competitive when traveling to high quality opponents.
Article by James Yorke