Preview · Premier League · 7 min read
The Cottage Has Never Been Kind, But United's Ledger Says It Rarely Has To Be
Fulham have spent three decades swinging between the divisions while Manchester United spent them stockpiling trophies, and yet the fixture keeps producing afternoons that neither record book can quite explain.
Rufus Okonkwo@thearchivist
Nigeria · The Archivist · September 20, 2026
Thirty Three Years Of Being Measured Against The Same Ruler
There is a particular cruelty in the way the Lemeister archive handles a fixture like this. Since 1993, Fulham have played 1394 recorded matches. They have won 524 of them, drawn 363, lost 507. Their goal ledger reads 1873 for and 1806 against, a surplus of 67 spread thinly across more than three decades of mostly honest, occasionally desperate, sometimes glorious football. It is a record that says: we turned up, we competed, we were rarely embarrassed.
Manchester United, over the same span, have played 1269 matches and won 754 of them. That is 59 per cent. They have drawn 278, lost 237, and outscored the opposition 2361 to 1250. A surplus of 1111 goals. Read those two columns side by side and you start to understand the shape of the thing. Fulham's entire modern history, the good bits and the bad bits, sits inside United's margin for error. That is not a dig at Fulham. It is simply what the ledger says.
Which is why Sunday at the Cottage, kickoff at half past three, is such an interesting test of a different proposition. The archive does not tell you who wins a particular afternoon. It tells you who usually has. Those are not the same question, and confusing them is how supporters end up furious at the wrong people.
The Long Arc, Told Through Two Columns
Let us take the numbers seriously for a moment, because they are starker than the rivalry's reputation suggests. United's 59 per cent win rate across 1269 matches is the mark of a club that spent most of this period as either champion or close enough to smell it. Six hundred and forty five of those matches did not end in defeat. Fulham, by contrast, won 38 per cent of theirs. Put another way: if you picked a random Fulham match from the archive and a random United match from the archive, you would be 21 percentage points more likely to be watching a United win. Over a large enough sample, that is not noise. That is climate.
And yet the goals columns are closer than the win columns in one revealing way. Fulham have scored 1873, United 2361. Across roughly the same number of matches, United's scoring advantage is 488 goals. Their defensive advantage is 556 goals. The gap, in other words, is not principally about United's attack overwhelming Fulham. It is about what happens at the other end. That is a quietly damning statistic for anyone who has watched Fulham ship soft goals at the worst possible moment, and it is a quietly flattering one for whoever has kept goal for United across these decades.
The archive spans 1993 to 2026 for both clubs, which means it captures the whole of the modern Premier League era and then some. It captures Fulham's yo-yo years, their Europa League run, the grim afternoons in the Championship and the happier ones since. It captures United's imperial phase, the Ferguson twilight, the long and expensive search for a successor's shape. Both clubs are in here in full. The ledger is not being selective. It is just being itself.
Model edge: draw +4.6 pts vs the market
A model probability, not a certainty. Analysis and education, not betting advice.
What The Model Makes Of It, And Why You Should Hold It Lightly
Lemeister's model has this one at Fulham 30 per cent, the draw 29 per cent, Manchester United 41 per cent. MeisterIQ, the conviction gauge, sits at 52 out of 100. That is a genuinely interesting reading. Fifty two is not conviction. Fifty two is a shrug with a spreadsheet attached.
Look at what the probabilities are actually saying. United are favourites, but only just. The draw, at 29, is very live. Fulham at 30 are not an afterthought, which is more than the 38 per cent historical win rate might have entitled them to. The model, in other words, thinks the venue and the current shape of both squads matter more than 33 years of ledger entries. That is usually sensible. History does not take penalties.
A forecast is a probability, never a promise. A 41 per cent United win means United fail to win 59 per cent of the time in the model's imagination. Anyone reading that as a guaranteed away day is reading it wrong. And anyone reading Fulham's 30 as a write off is reading the archive selectively, because the archive's whole point is that Fulham keep turning up, keep scoring, keep making a nuisance of themselves across 1394 recorded matches. Thirty eight per cent is not a doormat. It is a club that wins more than a third of everything it plays.
The Cottage, And Why The Run Of Play Matters
Craven Cottage is a peculiar place to play football. It is tight, it is old, it sits on the river and it has a habit of making visiting sides look like they have been asked to perform in someone's front room. The archive does not break these matches down by venue in the facts I have here, so I will not pretend it does. What I can say is that the historic spread of this fixture, United's dominance in the aggregate and Fulham's stubborn refusal to be a footnote, tends to compress when the ground is small and the crowd is close.
Tactically, the interesting question on Sunday is who controls the middle third and how quickly the game gets vertical. United's historical edge has been built on converting pressure into goals at a rate Fulham have never matched. Fulham's edge, when they have had one, has come from making the match scrappy, turning it into a series of duels and set pieces and half chances, and trusting that one of them lands. If United impose rhythm, the 41 per cent starts to look conservative. If Fulham make it a fight, the 29 per cent draw suddenly looks like the smartest number on the board.
And the draw at 29 per cent is worth pausing on. That is a high draw probability by any standard, and it reflects the model's sense that these two are not as far apart as the all time numbers suggest. Thirty three years of history tells you United win. The current model tells you it is closer to a coin flip with a lean. Somewhere in that gap is where football actually happens.
| Side | P (W-D-L) | Win rate | GF-GA | |
|---|---|---|---|---|
| Fulham | 1394 | 524-363-507 | 38% | 1873-1806 |
| Manchester United | 1269 | 754-278-237 | 59% | 2361-1250 |
Where The Ledger Goes Quiet
Here is the honest bit, the bit the archive cannot help me with. The totals I have are club totals, not head to head totals. I do not have the specific record of Fulham versus Manchester United across these 1394 and 1269 matches, and I am not about to invent one. The ledger, on this particular question, is quiet. What it does hold is the shape of both clubs: Fulham's steady, scrapping 38 per cent, United's withering 59, and the near symmetry of the fixture count, 1394 and 1269, which tells you these are two clubs that have shared a division often enough to have grown familiar with each other.
Familiarity in football does not breed contempt so much as it breeds precedent, and precedent is a trap. It tells you what happened. It says nothing about what will. A side that has won 59 per cent of its matches over 33 years still has to win this one on Sunday at half past three, in the rain or the sun, against eleven men who have their own reasons.
The Verdict The Archive Refuses To Give
So what does the record actually tell us about Sunday? It tells us United arrive with the weight of a far superior history, 59 per cent against 38, a goal difference of plus 1111 against plus 67. It tells us Fulham have spent 1394 matches being exactly the kind of club that makes a fixture like this uncomfortable rather than easy. It tells us the model, which is not sentimental, sees a 41 per cent United win, a 29 per cent draw and a 30 per cent Fulham win, with conviction sitting at a distinctly non committal 52.
That adds up to an afternoon that the past cannot settle. The archive has United ahead, decisively, across three decades. But the archive is a rear view mirror, and Sunday's match is coming at us through the windscreen. If Fulham keep it tight, if the Cottage does its usual work on visiting nerves, the 30 per cent starts to breathe. If United find their rhythm early, the 41 per cent will look like the bargain of the weekend.
Either way, the ledger will have one more line added to it by quarter past five. One thousand three hundred and ninety five matches for Fulham, one thousand two hundred and seventy for United. The numbers will shift, fractionally, the way they always do. And the argument about which of them means more will carry on exactly as it has since 1993, unresolved, which is the only way a proper fixture ever really ends.
- 1B. Lecomte
- 2T. Castagne
- 3J. Andersen
- 4Jorge Cuenca
- 5A. Robinson
- 6H. Reed
- 7H. Larsson
- 8Kevin
- 9E. Smith Rowe
- 10César Palacios
- 11Rodrigo Muniz
- 1K. Darlow
- 2Diogo Dalot
- 3L. Yoro
- 4A. Heaven
- 5N. Mazraoui
- 6Y. Tielemans
- 7Andrey Santos
- 8S. Lacey
- 9M. Mount
- 10M. Rashford
- 11B. Šeško
Premier League · Sun, 20 Sep 2026 15:30
Fulham v Manchester United
