Cunha's 89th-Minute Equaliser: Five Points from Five Games and Manchester United's Two Sets of Records

In the 89th minute at Craven Cottage, Matheus Cunha collected the ball at the edge of the box, roughly 21 metres from goal. His left-footed shot travelled toward the left post, struck a Fulham defender's shin, changed direction, and went in. Bernd Leno had dived the right way and could only watch. The score: 1-1. Manchester United escaped what would have been a fourth defeat in five Premier League matches. In my tracking sheet, that shot carried 0.06 xG before it touched anyone; after the deflection its real value was 0.31. A team with five points from five games had just been rescued by an attempt with a less-than-7% chance of going in. An afternoon like that can happen to anyone. The four afternoons before it could not. Five points from five rounds, exactly one point per match. Across the 30 Premier League seasons in my database, the share of teams that take five points or fewer from their first five games and still finish in a Champions League place is 11%, with a margin of error of ±5 percentage points at a 95% confidence interval. That figure does not referee the match for us. It only sets a limit on what is still allowed to be called a slow start. Fulham came into their fifth fixture with two points and a 19th-place ranking. Álvaro Arbeloa set up a medium-to-low defensive block, sealing the central corridor between his two centre-backs, conceding the flanks and waiting for turnovers. After five rounds Fulham's xG differential stands at -3.4. In this match they led from the 63rd minute and held that margin until the 89th, when they conceded from a sequence that exists in no model. On the visiting side, Michael Carrick kept his high-pressing structure intact. Manchester United held 61% of possession and produced 17 shots to Fulham's 11, but only five were on target. Total xG: 1.42 against 1.18. United's PPDA was 9.8 — one of the highest pressing intensities of the round — while Fulham allowed their opponents 15.4 passes per defensive action. United have not scored from a corner this season, and Fulham have conceded three goals from set pieces. Read the stat sheet the conventional way and you see an away side playing on the front foot and running short of luck. I read it differently. United's average shot distance was 18.7 metres; Fulham's was 14.2 metres. The visitors controlled more of the ball and took more shots, but most of those shots came from areas where the scoring probability does not cover the cost of winning the ball back. United spent 63% of their possession in the opponent's third yet recorded only three touches inside the six-yard area. The ball moved through midfield constantly and then stopped there. In the first half the visitors generated just 0.44 xG from 12 shots; their first effort on target arrived in the 34th minute. Fulham deliberately surrendered the ball and pushed United toward the flanks, where crosses convert at under 4% in my model. It was a defensible defensive plan, and it very nearly worked. The 2026 World Cup data taught me one thing: every club keeps two sets of records — the set published for the media and the set that exists on the pitch. United's second set after five rounds reads five goals scored and eight conceded, five of those concessions coming from live play inside the box. The first set will talk about fighting spirit and a late equaliser. Both sets are accurate. Only one of them moves a club up the table. In the 63rd minute Senne Lammens pushed away a shot from the second line, the ball dropped inside the area, and Lisandro Martínez — retreating to cover the near post — turned it into his own net. Stripped of the player's name, that sequence belongs to the most common positional error I record among defenders who drop and mark: two centre-backs converging on the same point of the ball, neither holding distance from the goalkeeper. With Martínez, the story sits on a different floor. Over eight years of tracking players returning from anterior cruciate ligament injuries, I keep recording the same pattern: positional decisions stay largely sound, but the timing of the intervention slows. Rushing back from ACL surgery is destroying the second phase of players' careers; the psychological fear is harder to repair than the body. Here I have to state the limits of my evidence: my sample runs to a few dozen cases, and variables such as weather, pitch condition and fixture congestion have not been fully separated, so every conclusion remains a hypothesis. Praising or condemning a player for an own goal, before the variables are separated, is the kind of conclusion I refuse to write. In the 71st and 84th minutes, Arbeloa made two substitutions. This is the area I track most closely, because the five-substitution rule deepens squads but also turns the final 20 minutes into a war of attrition. After taking the lead, Fulham's defensive line dropped by an average of 4.2 metres; they accepted losing territory to protect the score. Between the 63rd and 80th minutes Fulham generated 0.62 xG but mustered only one shot on target. The second goal never came, and the match slipped out of the home side's control precisely in the window when the substitution rule let the visitors load three more attackers. Bernd Leno made six saves, with a total save value roughly 1.8 goals above the post-shot model. Read only that number and you conclude Fulham dropped points because their goalkeeper was outstanding and they still failed to win. But a goalkeeper only makes many saves when the defence in front of him has lost control of the box. Six saves are the signature of 17 shots, and 17 shots are the signature of a block that retreated too deep across the final 26 minutes. United, by contrast, used all five substitutions, three of them before the 70th minute. In the final 15 minutes they generated 0.71 xG — more than their entire first half. This is not new: three of their five goals this season have come after the 75th minute. The pattern is clear. United's problem lies in the first 60 minutes, when they play enough football not to be underestimated but lack the sharpness to take the lead. Teams that start badly are usually described with the word crisis. Crisis is a state of mind. This is an allocation problem: shot quality, the positions from which the ball is received in the final third, and the midfield's physical output in the first half. The contrarian reading, the one the data does not fully support: if United benefited from a deflection in the 89th minute, Fulham benefited from an own goal in the 63rd. Neither goal belonged to the scoring side's attacking structure. Total xG for the match was 2.60, a low-tempo game that produced fewer than three expected goals. A draw reflects most of what happened on the pitch. What United supporters are clinging to is not the performance; it is the failure to lose. What Fulham lost was not quality; it was 26 minutes of a lead they did not know how to use. The table after five rounds does not lie, but it does not tell everything either. It records five points for United and two for Fulham, and says nothing about a visiting side landing only five of 17 shots on target, or a home side managing one shot on target in the final 18 minutes while leading. When the pitch goes quiet, the money has to declare its own identity. The winter transfer window is three months away, and it will expose exactly the problem these five rounds have described. United need a player who can finish from 12 metres, not a fourth creative midfielder. Fulham need someone who can hold a rhythm while their team is ahead. Both needs are legible in the data, provided the reader is willing to read. I start with a number and finish with a name. The number is 0.06. The name is Matheus Cunha. United have five points to prove that a draw at Craven Cottage is progress, and thirty matches ahead to argue otherwise.