19 September 2026
Brighton didn’t out-press Arsenal.
Arsenal just couldn’t get out.
Arsenal came to the Amex having won four from four and conceded once. They left having started 54 build-ups from their own third and finished one of them with a shot. This is what Brighton’s high press actually did to them, and, just as importantly, what it didn’t do.
Fifty-four attempts, one shot
Every time Arsenal restarted play from their own third, this is how far they got. A build-up counts from their first settled touch in their own third until they lose the ball, shoot, or the ball goes dead.
The read
The dashed lines are Arsenal doing this to themselves everywhere else. At their own rate across the other 42 league matches in this sample, those 54 build-ups should have carried them into the middle third about 30 times, into the final third 14 times, and produced between three and four shots. They got 24, 9 and one.
The cleanest way to say it: Arsenal normally start about 36 build-ups a match and arrive in the final third from roughly nine of them. Here they started 54, half as many again, and still arrived nine times. Every extra attempt bought them nothing.
They kept the ball. They just kept it in the wrong place.
Arsenal had 31 minutes on the ball, almost exactly their season norm. What changed was where those minutes were spent. Counting passes played out of their own third per minute of possession strips out the “they had more of the ball” explanation.
Own-third passes per minute of possession
White tick = Arsenal’s 2025/26 average of . This was their -highest of league matches across the two seasons.
Build-ups that reached the final third
White tick = 2025/26 average of %. Brighton held Arsenal to their lowest figure of this season.
The read
Possession volume was normal; possession territory was not. Arsenal passed the ball around their own third at nearly double their usual rate and turned almost none of it into progress. They weren’t starved of the ball. They were given it, in the place Brighton wanted them to have it.
Loose on the defenders, tight on everyone ahead of them
From tracking, across frames with Arsenal settled on the ball in their own third: how much space each Arsenal line was given, measured as the mean distance to the nearest Brighton player.
Space allowed, by Arsenal line
Higher = more space. Brighton gave Arsenal’s centre-backs and full-backs more room than anyone this season, and squeezed the forwards harder than anyone.
Who received Arsenal’s passes out of their own third
Share of all open-play passes Arsenal played out of their own third.
Did Brighton do this to everyone? Space allowed to each opponent’s back line
Mean distance to the nearest Brighton player while that opponent had the ball settled in their own third, all five of Brighton’s league matches this season. Chelsea played no nominal full-backs and are excluded from that row.
The read
This is a deliberate trade, not a failure to press. Brighton were happy for Gabriel and Konsa to hold the ball. They gave them a metre more room than Arsenal’s previous four opponents, and gave the full-backs a metre more too. What they squeezed was everything in front: the forwards were marked tighter than by any other Arsenal opponent this season, with the centre-backs stepping out man-to-man to do it. Struijk tracked Havertz at an average of 4.2 m for the whole match, Vuskovic sat on Gyökeres at 3.7 m.
The holding midfielders’ share of Arsenal’s exits out of their own third halved, from 22.8% to 12.5%, and the ball went back to David Raya more often than in any other match this season. Note what did not happen: Brighton did not mark Zubimendi and Rice any tighter than other teams do (they actually stood 0.3 m further off) and they screened the passing lane into them no more than usual. The pivot wasn’t smothered. It was made pointless: there was nothing beyond it to play to, so Arsenal stopped using it.
And this was built for Arsenal. The bottom panel is the giveaway. Against Aston Villa, Chelsea, Leeds and Coventry, Brighton pressed the opposition centre-backs at 4.3 to 5.1 m, close enough to make them hurry. Against Arsenal they stood 6.9 m off, over two metres further than their season norm and the loosest they have been on any back line all year, and gave the full-backs a metre and a third more too. They did not decide to press Arsenal harder. They decided to press Arsenal’s defenders less, and spend those bodies further up.
They could still break the first line. It led nowhere.
A line-break here is a completed open-play pass out of the own third, under 30 m, that goes past at least two Brighton outfielders standing in its corridor. Long balls are excluded on purpose: hitting it over everyone is not the same skill.
Line-breaks as a share of passes out of the own third
In raw count Arsenal played 22 of them, more than in any other match this season, because they had so many more passes to play.
Build-ups that reached the final third
Counts are small; hover for n. The gap between the two bars is the point, not the exact percentages. This measure is stable under a two-second shift of the event clock against tracking, which is why it is reported and a per-moment pressing metric is not.
The read
This is the part that surprised us. Arsenal broke Brighton’s first line at almost exactly their normal rate: 12.0% of their passes out of the back, against 10.2% to 15.7% in the four matches before. The press did not stop them playing through it.
What died was everything else. Of the 39 build-ups that contained no line-break, only five reached the final third: 13%, against roughly 26% in the previous four matches. Those are the possessions that normally progress by carrying out, combining down a side, or patiently working an overload. Against Brighton they were worth nothing.
So Arsenal were left with one working route out and no alternative. Fifteen of the 54 build-ups carried a line-break and four of those reached the final third; the other 39 produced five between them. When your only exit is the hardest pass on the pitch, a bad afternoon becomes a 3-0.
Where they hunted, where Arsenal died
Brighton pressing runs and high regains
Arsenal’s own half, attacking upwards. Brighton made 33 pressing runs, three times as many as Arsenal’s previous four opponents averaged, though Arsenal also gave them far more of the ball to chase.
Where Arsenal lost the build-up
Arsenal’s own half. 25 of the 54 build-ups died here, level with the worst of the last two seasons.
Brighton’s pressing runs, by player
Maxim De Cuyper drove it from the left: 9 of Brighton’s 33 pressing runs, and 5 of the 9 balls won inside Arsenal’s own third.
The long ball was no way out
Cut off through the middle, Arsenal went over the top. It was close to a coin flip, and they lost the flip more often than they won it.
Arsenal goal kicks played long
White tick = 2025/26 average of .
Long balls out of their own third
The read
Arsenal hit of their goal kicks long, far above their norm, and completed barely half of them. Across all long passes out of the back, Arsenal won the second ball less than half the time. The escape route Brighton left open was the one Brighton were set up to win.
The press didn’t change. Arsenal’s answer to it got worse.
Brighton have been pressing Arsenal like this for a year. The last three meetings, with Arsenal building each time.
The read
Read the pressing-run rows carefully, because they say the opposite of what you would expect. Brighton pressed Arsenal harder in December: 43 runs at 1.41 per minute of Arsenal possession, against 33 at 1.06 in September, and lost 2-1. In March they pressed least of all and lost 1-0. The intensity of the press is not what separates these three afternoons.
What has moved, steadily and in one direction, is Arsenal’s answer to it. Build-ups needed: 35, then 43, then 54. Reaching the final third: 25.7%, 20.9%, 16.7%. And this time the pressure finally converted into the ball: Brighton won it inside Arsenal’s own third nine times, against three and two in the previous meetings.
One thing was distinctly September: the press was front-loaded and led from a different side. Twenty-nine of the 33 runs came in the first hour, before the third goal, and Maxim De Cuyper drove it from the left. In March the chasing had been one man’s job. Georginio Rutter made 10 of the 21 runs on his own.
And yet the press didn’t score the goals
The honest part. Brighton won the ball inside Arsenal’s own third nine times and generated zero shots from those turnovers.
Every shot, by minute and xG
Circle size = xG.
Where the three goals actually came from
30′ Groß. A Brighton goal kick that didn’t stick, won straight back off Tzolis, then an eleven-pass settled move finished from the edge. Not a press turnover, a possession goal.
44′ Kostoulas. Arsenal’s own counter-attack: Ødegaard blocked, Saka saved. Verbruggen caught it, carried, launched, and Brighton were in behind three passes later. A goalkeeper counter off an Arsenal attack.
56′ Baixauli. A corner.
Brighton finished 1.70 xG as three goals; Arsenal turned 1.24 into none. On chance quality this was far closer than 3-0. The press’s contribution was not turnovers into goals. It was territory, field position and a set-piece platform, plus an Arsenal side that never got near Brighton’s box in settled play.
Was it just the scoreline?
A fair challenge: Arsenal were chasing from the half-hour. But the strangle was there before the first goal.
Build-ups reaching the final third, by game state
White tick = Arsenal’s 2025/26 average of 25.6%. The middle two spells are short (five and seven build-ups) so read the first and last rows.
The read
At 0-0, with the game flat, Arsenal were already reaching the final third from only 22% of their build-ups and losing 44% of them inside their own half, both worse than their season norm. The scoreline made it uglier. It did not cause it.
The lesson for Arsenal: the problem was never the first pass. Gabriel and Konsa had more time than usual. It was the second one. Against a press that man-marks the line ahead of the ball and lets the centre-backs dwell, someone has to break the marking: a pivot dropping into the back line, a full-back inverting, a centre-back carrying past the first line rather than passing around it. Arsenal did none of those often enough, went back to the keeper, and went long into a duel Brighton were set up to win.
The lesson for everyone else: Brighton’s press is not fast, it is disciplined. Their pressing-run rate per minute of opponent possession was actually lower than in their previous four matches. They didn’t run more. They ran together, in the same direction, for an hour.