Brighton won 1-0 at the Etihad on 31 October 2026. Yasin Ayari scored after a free-kick move, then a compact block turned City's 21 shots into noise.
Manchester City had 21 shots, 67 touches in the Brighton box and a goalkeeper in front of them who made seven saves. They still lost 1-0 at the Etihad on Saturday 31 October 2026, and the scoreline says more about how Brighton defend than how City attack.
Yasin Ayari scored in the 11th minute. The Guardian describes a scruffy sequence, with Sky confirming the Nunes error: a free-kick (the Guardian thought it was wrongly given against City's Anderson), Lewis Dunk sweeping the dead ball right to Diego Gómez, a cross, Matheus Nunes missing his clearance, Maxim De Cuyper keeping the ball in play and Ayari finishing into an empty goal with Gianluigi Donnarumma stranded.
The idea that decided it
Brighton treated a stoppage as an attack. Dunk swept the free-kick out to the right and the Guardian's Jamie Jackson says City "slumbered" from that point. It is a small coaching detail with a large payoff. Sky's numbers give Brighton 0.52 expected goals from set plays and 0.48 from open play, so just over half of their threat all afternoon came from set plays.
Both reports credit Brighton with good spells on the ball, but the rest of the job, above all after the break, was a low, crowded defence. Brighton made 15 interceptions to City's five, 32 clearances to 14 and six blocks to one. Eight of City's 21 shots were blocked before they reached Bart Verbruggen. City finished with 1.92 xG, Brighton with 1.00, but look at the price of each chance: about 0.09 xG per City shot and about 0.14 per Brighton shot. Volume without quality is the oldest trap in football.
The number
Sixty-seven to nine. That is City's touches in the opposition box against Brighton's, according to Sky's match stats. City also finished with 62.4% possession, 8 corners to 2 and five big chances missed. Iliman Ndiaye saw a shot tipped onto the bar, Erling Haaland went close from a corner and Nico O'Reilly, on as a substitute, headed wide late on.
None of that is mysterious. It is what a team looks like when it dominates the ball and cannot find a clean shot against a crowded box.
Same team, different job
The oddity is that Brighton normally press. Using Understat's per-match data, their PPDA (passes the opposition make per defensive action) is 7.19 over the first nine league games, the lowest in the division. Here they had 37.6% of the ball, made 32 clearances and their PPDA for the match rose to 11.7 on the same data. Fabian Hürzeler's side looked to change the plan to suit the opponent, which is rarer than it sounds.
For City it is a first back-to-back league defeat under Enzo Maresca, a week after the 2-1 loss at Aston Villa on Saturday 24 October. Maresca dropped Marc Guéhi and Rayan Cherki for Nunes and Ayyoub Bouaddi, then brought Cherki on in the 55th minute. It did not change the outcome, but the Guardian was generous about a team it called "slick throughout".
What it means for the table
Not much panic is needed at the Etihad. City sit top on 21 points from nine games, level with Arsenal, who have a game in hand, and Understat's expected points model has them on 19.1, so the table is close to fair. Brighton are third on 19 points, two behind, and Understat has them on 15.9, so they are running ahead of their numbers.
That is the bit to watch. On the day Brighton scored once from 1.00 xG on Sky's numbers, which is exactly par, though City's 1.92 says they should have scored at least once. It was also their first win at the Etihad, per Sky. Whether it makes them a top-four side or just a very well-drilled one is a question for the next nine games.
Golden Era's AI newsroom wrote this from the sources below, and an AI standards desk checked every claim against them before it went live. No human read it before it was published. A human editor oversees the newsroom and corrects mistakes when they are found. Shiv Kearney is a fictional AI reporter. The picture is an AI-generated illustration. How Golden Era works














