Manchester United Fail to Beat Fulham: Carrick on Two Woodwork Hits and Leno's Display

Bernd Leno was named Player of the Match. For a goalkeeper, that award usually arrives in one scenario: the opposing attack keeps pouring forward, the defence accepts sitting deep, and the man in goal becomes the last checkpoint. In the match where Manchester United failed to beat Fulham on Sunday 20 September 2026, two United efforts found the woodwork, and one phrase was repeated after the final whistle: "deserved three points". Michael Carrick did not call it a tactical failure. He called it a performance failure. "We created quite a lot of chances to win this match. We played well, and we were not lucky enough to concede in that way," Carrick said. "Their goalkeeper was outstanding today. He made many saves... Today two of our chances also hit the goalpost." The record I have in hand contains four propositions: Manchester United failed to win against Fulham; Carrick assessed that his team deserved three points; Leno was named Player of the Match with "many saves"; and two United chances struck the frame. The image is credited to AP Photo/Kin Cheung. The original publishing outlet is not named. What is absent matters just as much. There is no scoreline. No starting lineups. No formation. No shot map. No xG. No actual save count for Leno. No league table, no positions for either side, and nothing to establish whether United dropped two points or three. One reasonable inference: Carrick speaking about three points as something taken away suggests United were rated the higher side before kick-off. That is an inference from a quote, not from data. I state that before analysing anything, because it is the habit of the trade. "Clear and obvious — the way sports law names its own helplessness." A football match technically leaves behind hundreds of data points. A four-proposition record leaves four. The gap between those two things is the gap between an event and a story told about an event. Based on my experience following matches, I split post-match claims into three layers. | Layer | Content | Verifiable? | |-------|---------|-------------| | Event layer | United did not win; Leno won Player of the Match | Yes, with official data | | Measurement layer | "Quite a lot of chances", "many saves", "two woodwork hits" | Only with a shot map | | Interpretation layer | "Deserved three points", "not lucky enough" | Not verifiable by any tool | Only the second layer decides the real story of the match, and it is also the layer with the least data. "Quite a lot of chances" is a category with no agreed definition. It could be four shots from outside the box, or three one-on-ones with the keeper. In a coach's decision model, those two sets lead to opposite conclusions: one is a finishing-selection problem, the other a structural attacking problem. The two woodwork hits are the most notable facts in the record and the easiest to misread. The post sits centimetres from the goal line. A shot a few centimetres inward is a goal confirmed by goal-line technology; a few centimetres outward is a line entry under "nearly". That margin is so small it says very little about shot quality and a great deal about the randomness of the final outcome. Opta recorded that the 2026/21 Premier League season ended with 40 penalties, the highest single-season tally in the competition's history, most of them arriving after IFAB amended the handball law. No defender became clumsier over one summer. What changed was the definition. Apply that logic to the Fulham match: if the threshold for calling a situation a "clear chance" shifts, the conclusion about the match shifts with it, even though the ball rolled exactly as it did. At the same time, that very randomness is why professional clubs build xG models — to separate chance quality from chance outcome. I still hold that xG is seriously overused and often turned into a judgment index rather than a descriptive tool. In this specific case, however, the absence of xG turns every conclusion into guesswork. A match in which a team creates "quite a lot of chances" and does not score can be evidence of a blunt attack, or evidence of a goalkeeper in an unbeatable state. Those hypotheses lead to different training-ground responses, and no data in the record distinguishes them. "The VAR machine does not blow the whistle; it only teaches us how to see what we are about to believe." Goal-line technology does exactly one job: establishing whether the ball crossed the line. It does not measure, and was not designed to measure, whether a team deserved its result. When a coach says "we deserved three points", he is operating on a layer no device can reach. The most plausible model behind the record is a low defensive block plus a goalkeeper at peak form. That model is real, and it can repeat across matches if a side accepts ceding territory to protect its points. But the record does not describe Fulham's defensive shape, so this remains a hypothesis. One part of Carrick's answer matters more to me than the chances. "The way we ended the match is a positive thing we can take from this match." That phrasing hints at a specific game state. If United needed a reaction after falling behind, or if United surrendered a lead, the central problem is no longer finishing efficiency but game management — a different skill, measured by the ability to keep the ball in the opponent's half in the final fifteen minutes rather than by shot count. Read that way, Carrick spending most of his airtime on Leno becomes a deliberate choice. Praising the opposing goalkeeper moves responsibility away from his own attack without defending anyone explicitly. A striker reading that interview hears the message: this result was not on you. That is a psychological management tool, and it works in the short term. It only stops working when it repeats for the fourth or fifth time in a season. "I do not watch matches through a spectator's eyes, but through the eyes of the man being judged by spectators." From there, the standard for judging a "should have won" match lies in repetition. One match lost to an outstanding opposing goalkeeper is an accident. Four in a month is a pattern. The initial record contains one match, so there is nothing yet to say about a pattern — and that is all that can be said honestly. What I want from Manchester United's next matches is not more shots. It is better data disclosure. Major leagues standardised per-match shot maps years ago; a coach talking about performance would be far stronger attaching chance-quality data rather than the adjective "quite a lot". And until then, remember that this match was decided by things no statistics table contains: a few centimetres of a goalpost, and a goalkeeper having an evening in which every reflex was right.