WTF happened against Brighton
An excavation of the humbling 3-0 loss, why it wasn’t as bad as you remember, and what it tells us about summer priorities
Fifteen minutes into a fitful and physical opening stanza against Brighton, Martin Ødegaard let out a speculative banger from distance:
It came oh-so-close to going in:
Did that shot look familiar?
Last week, in the exact same period of the game — 13 minutes in, though in a much more negative run of play for Arsenal — the skipper blasted a carbon copy into the corner webbing. It provided a lead that was never relinquished:
Back to this Sunday. Eight minutes after that narrow miss, Gabriel Jesus took on Lewis Dunk 1v1, beat him to the byline, and ripped a powerful shot that gave Jason Steele some trouble. It went out for a corner:
This was a mirror image of Martinelli’s opportunity last week — which was better defended, less powerfully struck, and not even aimed at the net. Of course, that one went in:
If this is starting to feel a little bit like an overly sanguine, shrug emoji, football-is-like-that-sometimes type of post after a tremendously disappointing result, that is not my intention. Or maybe it is — that is certainly part of the story. But there’s more to learn from this one, so let’s dive in.
To start, Arsenal had no shortage of good opportunities, particularly in the first half. The reason? The press.
Chance-creation via press
There’s an interview with Jamie Foxx where he was asked what it was like to work with Quentin Tarantino on Django Unchained:
“He was a tyrant. That’s what you want. You want a director that if, even if you’re going off the cliff, you know you’re going off the fucking cliff.”
I was reminded of that quote when I saw a profile on Roberto de Zerbi, where he shared his philosophy:
"I can decide how I can lose, not how I can win. I want to lose playing with this style, this courage, to try to command the game, the play. Sometimes you can lose, but it's my style, my idea, and the players are with me."
In the Brighton build-up phase, both teams played perfectly to type.
Brighton preferred their usual, swaggeringly patient 2-4-4 approach, with the keeper orchestrating the build-up to gain numerical superiority. The wingers were pinned way high and wide, sometimes swapping sides (which is something I want Arsenal wingers to do occasionally, btw), way out of frame, and Jesus would swing around to direct play to the Brighton left:
It must be said, the Arsenal press looked active and dangerous from the jump — with Jesus hounding them out of play on the first possession.
In all, Arsenal was able to generate 74 recoveries, including 31 in the middle third, and 21 in the attacking third (compared to 9 high recoveries from Brighton).
In total, Brighton had 29 low losses.
Six of their losses led directly to Arsenal shots. Let’s go through a few of them.
As we started this post with a few almosts, there was this narrow miss on an interception by Xhaka, which could have been a walk-in goal:
That Jesus shot above? It originated with Colwill trying to break the lines:
…and Jorginho appropriately hounding Mac Allister into a dispossession:
Seven minutes later, the team absolutely enveloped Enciso as he tried to dribble out of pressure:
Once the ball was loose, Kiwior didn’t need to control it — he tapped it directly to play Ødegaard into space. These one-touch-passes-from-dispossessions are my sauce:
That set off the events that led to a Trossard rip off the crossbar. He used Tierney’s overlap to generate the necessary space to get the shot off, but may have been better served playing it to the left-back. The Scot certainly thought so:
A little later, Trossard himself nabbed it upfield, and there was a near-perfect Jesus cross into the box, where there were ample numbers:
But as Colwill fell down, making it an even easier opportunity, Ødegaard couldn’t reach it with his head, and it landed in an awkward place for Saka to get enough leverage for a shot off. Nobody was really at fault here; it was just one of those things.
But as a reminder when looking at stats, a play like this results in 0 goals, 0 shots, and 0 xG:
There was also this powerfully-struck Saka hit, with a beautifully slow wind-up that just sailed it wide:
Before the half, Tierney and Xhaka swarmed Enciso on the touchline:
…and got it up to Ødegaard for a shot across the mouth of the goal:
There were other opportunities, but all this pressing activity meant that Brighton had to look elsewhere for chances. That “elsewhere” was long and over-the-top.
But we’ll get back to that later.
I am the passing jinx
It’s not quite as simple as some unfortunate, narrowly-missed chances, however. There’s more structural responsibility on the players.
Before we go any further, I’d like to apologize. Last week, I wrote this:
“But when are Arsenal at their best? When they are plucky, direct, and take measured risks. As an example, here is a list of Arsenal’s ten lowest-percentage passing games of the year. Notice anything? That’s right: 9 wins, 1 draw, no losses.”
You’ll never guess what happens next:
The conclusion here is not “in order to win, pass at a low percentage.” It’s more nuanced than that. In its current form, Arsenal is best on the front foot, and can have a little trouble in lower gears; they are not equally effective across tempos. They are most dynamic when they play vertical and opportunistic.
That was clearly the plan against Brighton. The plan was to chuck possession and pure efficiency out the window, in lieu of creating tangible chances while Brighton were spread out in their man-marking scheme.
Why wasn’t this attack crisp enough to score goals? I have a few thoughts.
1. A lonely Jorginho
The single biggest problem with Arsenal’s build-up was the illusion that it occurred. Keeping up with a direct gameplan, Ramsdale bypassed phases and went long, often directly to Jesus:
As we covered in the West Ham recap, Tierney then had a negative impact on build-up as a Zinchenkoan double-pivot player. Before this one, I wrote that I’d like to “Let Tierney be Tierney” and perhaps roll with a Jorginho (LCM) and Thomas (RCM) double-pivot, and let Tierney bomb down the line, instead of the slower Xhaka/Jorginho pairing, and the misprofiled Tierney in midfield. I still wonder if that may have been the right call.
Nonetheless, Tierney occupied more of a prototypical left-back role in build-up, hugging the touchline without exception. He didn’t really put a foot wrong and did a wonderful job of zeroing out the attacking threat on his side, but was notably uninvolved and unassertive in the passing game, completing only 19 passes. (For reference, Zinchenko completes about 60 per 90).
To make up for this, it was often Ødegaard dropping in to form a double-pivot:
This had a few knock-on effects: though Ødegaard dropped deep, he didn’t get the ball very often in that position. And when lines were broken, he was typically behind the ball instead of ahead of it, where he often does his best work (he still made due). Lastly, this gave Jorginho a lot of room to cover both in- and out-of-possession, and his legs are too similar to mine for that to be a great idea.
This one doesn’t take a Twitter Tactico to figure out. The aggressive passing was a little less crisp than usual because there was a Zinchenko-shaped hole in build-up.
2. Fouls and interruptions
There’s an easily-observable hack of Premier League refereeing that plays out almost every week against Saka (and others): you can get away with whatever you want for the first 20 minutes or so.
The game started with Ødegaard getting pucked in the face with a ball (not a foul, obvs), escalated to a cardable Martinelli shoulder on Mitoma, and really got in its groove with Caicedo’s unpunished roll of Martinelli’s ankle. Whereas everybody was expecting a clinic in beautifully flowing football from both sides, the game got surprisingly physical — and oft-interrupted.
There seems to be an appreciable effect of these interruptions on Arsenal’s crispness in delivery and final actions; it also seems to impede their ability to settle down in advanced possession and confuse the opponent with rotations, as we have seen so often. To note:
If Arsenal is fouled 10 or fewer times, they have 0 losses
If Arsenal is fouled more than 10 times, they have 5 losses
This is not unique to our boys, or free-flowing sides in general. In the five games that Man City have been fouled the most, they are 1-2-2.
Somewhere, a single tear rolls down Sean Dyche’s cheek.
3. The midfield carries a deficiency
Carries a deficiency, get it?
I may do a piece on why Arsenal are still behind Man City, looking past the obvious (they have nation-state money and nation-state depth) and into the nuanced facets of the game in which there is still a gap.
One of the first things on my mind is that Arsenal can be overly-reliant on its wingers to carry the ball up, which is fine in most cases, but not the best path in every situation. Martinelli and Saka frequently lead the league in progressive carries, but those in the middle of the pitch tend to fall behind as an outlet when passing won’t quite do the trick.
Here’s a table of carrying data that compares the stats of the most-used Arsenal midfield trio to that of Man City:
This is particularly an issue with Xhaka, who is often presented with opportunities to infiltrate the press with a carry, but isn’t agile enough to do it. This may have been decisive against Liverpool. It reared its head again versus a similar-intensity side on Sunday.
Billy Gilmour, on the touchline, traveled 10 yards to take this off Xhaka on the latter’s first touch in a transition opportunity:
Here, he was fouled before getting the ball away in this transition:
…and here’s the kind of play where Xhaka demonstrates all the potential of the role, without having the level of athleticism to exploit it.
He manipulates space and sneaks behind defenders to get the ball in green grass. But from here, the issue is that no defenders feel the need to go up and commit to him, because they’re not concerned about him making his way to the box. They sit back in their comfortable zones and basically leave him unmarked:
…and now that Caicedo has squeezed out enough help, he takes one step forward and intercepts it. Attack over:
Improving the quality of our midfield carries is a tangible way to upgrade this summer.
Basmati, anyone?
4. Lulls in keeper-testing
Typified by the new shooting boots of its captain, Arsenal is not shy to pull the trigger this year. They lead the league in total shots, and are fairly efficient at converting overall opportunities: their .13 goals per shot in second in the league, behind only Man City.
The issue is that this relative clinicism is not equally distributed amongst its games. For all the hand-wringing about this one, you’re not going to win a lot with only 2 shots on target.
The team has so far logged seven different games with 2 or fewer shots on target, and they’re 3-0-4 in those. Comparatively, Man City have only two such games, and are better at generating penalties regardless.
What’s the fix? Getting the ball on the net more consistently and seeing what happens. Arsenal pumps a lot of shots in, but they are only 13th in the league at getting those shots on target; though they score at a nice clip, they don’t trouble the keeper relentlessly enough across games.
Gone are the days of “Arsenal always tries to walk it in.” Enter: “Arsenal always tries to graze the post.”
It’s the transitive property of shooting. More shots on the keeper = more mayhem = more rebounds = more goals.
What about the defending?
I’m afraid I don’t have particularly novel analysis on this one. The press was rocking for most of the day. It tapered off some in the second half, allowing Brighton to play through in a few different ways — but there were still 9 high recoveries, and a similar amount of duels won and shots taken. To my eyes, this was not a Man City redux of “6-man build-up demolishes Arsenal press.”
Though Colwill held a tremendous amount of volume and made many high-quality passes, he also had 13 losses, many of them in dangerous areas:
With Tierney neutralizing the Brighton right, and Gabriel successfully accompanying Evan Ferguson like a trusty sentinel, the Brighton gameplan targeted Benjamin White in space.
This was often through low-percentage opportunities over the top, and I don’t think it’s the kind of ball that Arteta would try to avoid schematically. I don’t think it’s an example of De Zerbi’s tactical genius, either. These balls are hard to place: Ramsdale and Kiwior are always lurking to clean ‘em up, and so is the touchline. If that all fails, White generally has enough pace to stay within a step or two, then slow things down for the calvary to arrive, which is all part of the plan.
On the day, though, the pressure on White was cumulative, and though he’d get himself back in good position, he was then flat-footed in winnable situations.
In Brighton’s best chance of the first half, Mitoma gathers the ball wide in an isolated situation that is acceptable from Arsenal’s perspective:
Within seconds, the entire Arsenal cohort is back in their desired positions, setting themselves up well. But two factors lead to a dangerous cross by Mitoma: White overcommits and gets turned, and Jorginho isn’t fast enough to cover the byline:
The first goal started back here, as most of Brighton’s play did. This view gives a really good sense of Brighton’s shape — including the 2-4 in build-up, and Enciso making runs across Ferguson. Colwill hit it over the top to Mitoma to get him running, and I highlight Estupiñán for a reason. This is where he started the overlapping run on this very play:
Mitoma must be shaded towards his right foot (he’s brilliant at cutting across zones and firing shots in), and leaves it to a dead-sprinting Estupiñán. Mitoma is like an NBA player performing a pick play, and Saka has to wind his run around him, which disrupts everything. At that point, my feeling is that White should have rotated down, or Jorginho should have recognized this and tracked the overlap — but I’m guessing he didn’t feel confident enough in his speed to make it happen:
It’s worth noting that Saka made several deep, successful, max-hustle defensive plays in the second half after this.
From there, Estupiñán got his step to cross it in. Several annoying things followed: Kiwior got cleated and went down; Tierney’s header went right back to the left-back; and Estupiñán somehow managed to nail it directly into the ground, over Tierney’s head, and right to Enciso, directly into the spot vacated by a fallen Kiwior. Bah:
(Another side note: I saw people raising issue with Kiwior staying down there, or not being savvy enough with his reaction. If that was you, congratulations: you have won the first Edu’s BBQ raffle! The grand prize is to have a giant Irishman come by your house and stomp on your achilles with his cleats!)
After Arteta’s post-game comments, I had an eagle eye for any signs of relative lethargy in play. I did notice that Ødegaard and Jesus never made it in the picture here, but my guess is that’s intentional as the leaders of the 4-4-2 press, and this still being an 7-on-8 with a keeper. But I note what I see, dear reader!
The next 30 minutes saw a lot of play that looked largely like the first half: an evenish affair, with spates of attacking brilliance that ultimately didn’t materialize. Then came a big turning point: the subs.
While I support the broadstrokes of the subs — namely, Trossard moving to left-8 while chasing goals, Reiss on the left instead of the right, and Nketiah getting a chance in an attacking mode — the ultimate lineup (from 77’ onwards) was without the likes of every one of Jorginho, Jesus, Xhaka, and Ødegaard: AKA, all of Arsenal’s most vocal, organizing players in attack.
What resulted was a bit of schoolyard, solid-effort, but less-organized play … and two more goals for Brighton. I’m all for Ødegaard being rested and rotated, but rarely see a benefit to subbing him late. He doesn’t tire, his assertive passes are often exactly what the doctor ordered, and his communication is often much-needed.
The next Brighton goal was the result of a too-cute mistake by Trossard, trying to flick it up to the 9 without looking:
When Trossard was signed, our post was largely a gush-fest. That said, after a honeymoon start to his time here, Gooners may have to adjust to a more nuanced portrait of his game at times:
“He’s not without warts: he struggles in medium-passing and can have stretches of games where his impact wanes. There’s little reason to think he’ll struggle due to tactics, as he’s shown a lot of flexibility over the years, but he may be prone to periods of lower form, and isn’t entering the team on the best run … There will be plenty to iron out: despite the shortened list of requirements in build-up, is he secure enough on the ball to make it work?”
The third goal was because Ramsdale didn’t hold on to a spurious long shot from Undav. It was a little disorganized, but Kiwior was up the pitch trying to disrupt play and force goals, which was exactly what he should have been doing.
What do we make of all this?
After the match, Arteta offered the fans an apology. For his players, he offered some of his sternest words yet:
“When you have to play in these moments you cannot do what we did in the second half. Then we have to look. If a team is capable of doing that when it comes to the biggest stage, there’s a lot of things to analyse and think about because it cannot happen.”
Ødegaard was similarly despondent, with Kiwior saying “our heads dropped” after the first goal.
Now, I don’t think Arsenal players are necessarily covering themselves in glory with their recent press hits, and I also trust them to have a firmer grasp on the team’s psychology than me. But because of these quotes, I rewatched with a keener eye towards overall effort than usual — and honestly? I didn’t find a ton to quibble with.
There were cascading issues that led to the first goal, all of them understandable and unfortunate in my book. The Trossard mistake was the stuff of coaching nightmares, and he may not have always been at his highest-effort throughout, but it did not seem to be indicative of a systemwide failure in composure (as I believe we saw at Anfield). Tierney didn’t seek to impose himself on the proceedings, even when things were desperate — which became clearer when he was in the same areas as Reiss, who clearly did. But overall, my impression of the game was more mixed than purely negative, and I left with fewer concerns about the structural integrity of the Arsenal direction, and more anxious to get the summer transfer window going.
Some thoughts on a few players I haven’t touched on at length yet:
I’ve been slightly concerned that an extension to Reiss Nelson fell into the sentimental, accountant-approved category rather than the “fuck it, let’s try to win the Champions League” category. But he has been genuinely electric in every week bar one, and it’s no longer a small sample size. He’s earned the right to do whatever he wants. Whatever that is: godspeed, sir.
I came away from this game about as high on Kiwior as I could have imagined being at this stage. His close-touch controls are elite, as is his recovery speed. His judgment of the game is developing but his first instincts are often spot-on, and I was particularly impressed with his one-touch passes after winning possession (which show a sense of anticipation). Surround him with communicators and watch him soar. I’m holding my stock.
Jesus was a dynamo of activity, and is hugely responsible for the level-setting of the press. In possession, he tended to overdribble, which led to a voluminous, inefficient game overall, with too few passes. I’m not sure he’s fully rediscovered his pre-injury form yet, but it’s not for lack of effort.
White was a big issue on the day, unfortunately. His level of focus and activity was at his normal levels, but it felt like his body was a step behind his mind in a few pivotal moments. Having a superior athlete to spell him — whether a purer RB, or a faster RCB hybrid like Simakan — will go a long way.
Saka wasn’t decisive enough about passing out of his double-teams, and wasn’t at his best — but wasn’t at his worst, either.
I have a theory about Jorginho that I’ve been meaning to verify: is he least effective in equal game-states? When down a goal or two, he can cheat forward and pick out assertive, manipulative passes. When up a goal or two, he can cheat backwards, dictate tempo, slow things down, and frustrate the opponent with his gamefeel. When it’s 0-0, he can have too many possibilities to cover.
And some final notes on tactics:
I would have preferred a double-pivot with Jorginho and Thomas, but ultimately, have little issue with Arteta’s overall tactical plan; it created a cavalcade of good chances in the first half, and the second half was not without opportunities: by Nelson, by Trossard, by Big Gabi (with the captain’s armband, praise be). He gave the players enough of a chance to win.
I do have some concern about the “kitchen sink” lineup lacking organization late in the game. There wasn’t enough structure for the whole to be confident and successful.
OK, onto the lessons.
What did we learn for next year?
This team is still great, but it needs to work on expanding its window of tolerance: in other words, how to establish dominance when things don’t neatly go to plan.
None of this is news, but it was reinforced on Sunday:
No, the press is not broken — Pep didn’t provide a one-size-fits-all battleplan for everyone to use; it was pretty specific to the abilities of Gündoğan, KDB, and Haaland, and particularly potent against a specific group of beleaguered or injured players. This hybrid pressing scheme is still capable of some great things, and will be nicely helped by the next bullet.
Midfield speed, physicality, and carrying ability — We saw Xhaka miss a walk-in goal by a step. We saw Jorginho miss defending a byline cut by a step. We saw defenders unbothered by their carries. These are brilliant players laying out a perfect blueprint for those with more physical gifts. The ideal Arsenal press and build-up offers a wide swath of real estate for midfielders to roam; with a little more speed and carrying ability in those positions, the marginal improvements could be enormous.
Rotation and speed at right-back — Depending on the matchup, this would be mighty nice, and Sunday reinforced the need for reinforcements.
Playing interrupted — The team needs to do a better job of maintaining its #vibes and precision when the game is interrupted, fouls be damned.
Shots on target — This team can, indeed, be clinical — but may try to be too clinical, too often, and would be better spraying the keeper with more shots on target.
More routes for progression — The team needs to continue to devise alternate routes of progression when Zinchenko is out (or otherwise neutralized). In the interim, Ødegaard dropping deep gave me painful Sambi/ESR/Ødegaard midfield vibes from last year at Anfield.
OK, that’s plenty.
It was a variance-heavy game, where the variance didn’t fall towards Arsenal; but Arsenal also missed clear opportunities and had more exploitable performances overall. There are tangible, attainable areas of improvement for the young side.
I hope that gave you a slightly more nuanced view of the game than “fuck that shit.” But if that’s still your view? Honestly, fair.
It’s been an exciting week for transfer reports, and I’ll be getting into that mode pretty quickly — with a Big Board, updated rankings, and more. In the meantime, I shared some thoughts on Zubimendi (unlikely) and Simakan (reported interest) over on reddit.
I appreciate you.
Happy grilling.
❤️
my friend this is a thoroughly good analysis - thank you