Man United 1-1 Fulham: 29 Shots, 1.6 xG and Lisandro Martinez's Repeating Error

In the 89th minute at Craven Cottage, Matheus Cunha collected the ball inside the box and equalised for Manchester United. The goal salvaged a point, but it did not erase the page of notes I kept open for the whole match: 29 shots, 13 corners, 606 passes at 90 percent accuracy, and 1.6 expected goals. A side that out-shot its opponent nearly three to one still had to wait until stoppage time to rescue a draw against a team that started the day bottom of the table. I watched this match on two screens, one showing the footage, the other a stats board updating with every attacking sequence. The gap between the two widened as the game went on, and that is why I am writing this.

After five rounds, Manchester United have five points and remain in the lower half of the table. The gap to fourth place is four points, which sounds tame, but it only means something if the team owns a stable mechanism for scoring. Fulham started bottom and played exactly the football most struggling Premier League sides choose against a big club: a deep defensive block, a midfield folded into two lines, space conceded in front of the box, and a willingness to live with crosses. Craven Cottage is small and the stands sit tight to the touchline, so a visiting side holding the ball tells us little about who actually controlled the game.

Before dissecting anything, the sources and the limits need stating. The numbers I use here come from public expected-goals models, cross-checked between two data providers, and I only kept figures that appeared consistently in both. xG is not truth; it is a probability estimate built on location, body shape and shot type. In a single match, model error is large enough that I have to verify against video before drawing conclusions. Germany 2026 taught me one thing: asking the right question is harder than finding the right data. That year my model gave Germany an 82 percent chance of escaping the group stage, and they finished bottom. The data did not lie, but it answered a different question from the one I thought I was asking.

The right question at Craven Cottage is not how many times the visitors shot, but how much those shots were worth. Twenty-nine attempts sounds impressive until you divide by 1.6 xG. That lands around 0.055 xG per shot, while the average for a major Premier League side usually sits between 0.09 and 0.11. Put differently, every time a United player pulled the trigger, he generated roughly half the value of an average shot in this league. Of the 29 attempts, 17 came inside the box and 12 from outside it, but that geographic split means little without body shape. Ten shots blocked and 13 off target is the familiar signature of shooting under pressure, when a defender is already in the blocking lane before the foot meets the ball. Only six were on target, and only two were classified by the model as genuinely dangerous chances.

Fulham, with roughly 42 percent of possession, produced three clear dangerous chances, more than the visitors. That is the anchor of this entire analysis. A deep-defending side does not naturally out-create a possession side unless the possession side hands it transition moments. And that is exactly what happened.

The conceded goal began with a build-up phase I rewatched four times. Kobbie Mainoo had the ball in front of his own box, facing his own goal, and chose to push it quickly up to Cunha in midfield. The trouble is that Cunha is not a genuine centre-forward with the strength to hold the ball with his back to goal. The Brazilian is lighter than Bassey in that duel and went down almost immediately. Cunha looked for the whistle, the referee waved play on, Bassey carried the ball into the box and shot. Lammens saved superbly, and the moment should have ended there.

What followed is the detail that made me stop the tape. Lisandro Martinez had both the time and the space to clear the ball into touch. Instead, the Argentine defender cushioned it back toward Lammens while the goalkeeper, who had just dived to the opposite corner, was scrambling back. A safe clearance does not produce a goal; a risky touch does. Across the first five rounds, Lisandro Martinez has now made direct errors leading to two Manchester United goals: the careless loss of possession against Ipswich Town and this moment. Two errors in five matches is not luck; it is a behavioural pattern. Both came from the same source: a belief that individual skill can solve a situation where safety is the optimal choice.

Midfield is a different story, and I think it is the tighter knot. Bruno Fernandes remains the soul of this team in central areas, but he is not good at manufacturing the combinations that stretch a deep block. Bruno is at his best with sudden line-breaking passes, a weapon that only pays off when space exists behind the defensive line, meaning in transition. When an opponent builds two defensive layers in front of the box, that space disappears, and his default becomes a cross or a shot from the edge of the area. Premier League defenders are coached to read exactly those two actions. Thirteen corners are the direct consequence: the ball kept being pushed wide after attacking sequences that offered no give-and-go option.

Based on my experience tracking Premier League matches, an attack that clears 25 shots without reaching 2.0 xG usually has a structural problem rather than a finishing problem. Here the structure shows up in the tempo. United's players know they are facing an eleven-man defensive block, yet they still dawdle while waiting for teammates to arrive and occupy positions before releasing the pass. Every such pause hands the Fulham back line another two or three seconds to rebuild its shape. That stuttering rhythm is not the product of one underperforming individual; it is the consequence of missing a third-man attacking option, the kind of combination a two-layer block cannot answer simply by dropping deeper.

At this point I have to argue against myself. The familiar shorthand of a blunt attack and a leaky defence bundles two very different problems into one phrase and blurs what should be fixed first. On this match's evidence, the defensive problem is not that severe: United conceded very few situations, and the goal came from an individual mistake rather than a systemic collapse. The attacking problem is the opposite, systemic and repeated week after week. Holding 58 percent of the ball and taking 29 shots did not cause the stalemate; both are the output of a third variable, namely the absence of any attacking plan beyond delivering the ball into the box by crossing. The possession share is also a consequence of Fulham ceding the ball on purpose, not proof of dominance. Anyone reading that 58 percent as a strength indicator is reading the causal arrow backwards.

An analysis that only presents evidence supporting its own thesis is worthless, so here is the counter-evidence. The three-to-two edge in clear dangerous chances is tiny, well inside the error band of a single match, so it cannot establish that Fulham deserved to win. The 89th-minute equaliser also shows that the visitors' fitness and mentality have not collapsed, an important fact five rounds into a season. And if the sample is widened across all five matches, United's total xG may still rank respectably in the league, something one game cannot disprove. This team's xG at Craven Cottage did not create a crisis; it simply showed that the crisis was already there.

Over the next three rounds I will track three numbers. First, xG per shot; if it does not clear 0.08, the problem lives in the attacking structure rather than in finishing form. Second, the count of cutback passes from wide areas into the box, the weapon a deep block struggles most to defend. Third, touches inside the box by the forwards, a gauge of whether Cunha and his teammates are receiving the ball close enough to goal. Those three numbers will answer what the 1-1 scoreline cannot: whether Carrick is fixing an individual error or redesigning an entire scoring mechanism. If it is the latter, the four-point gap to fourth place stops being a comfortable figure and becomes a distance measured in the rounds left on the calendar.