Foden is suspended, and City have never looked stronger
Sent off in the derby, the Englishman misses Sunderland's visit. Our model gives Manchester City 71% without him — and has them going backwards if he came in. This is not a charge against Foden; it is a survey of the area he used to own.
By Lucas da Silva · September 18, 2026 · 5 min read
Some absences show up on a teamsheet. Others show up nowhere at all. Phil Foden's, on Sunday, belongs to the second kind.
Sent off in the derby at Old Trafford — which City won with ten men — the England international serves his suspension. This is the moment where you write that Manchester City will be weakened.
Our numbers say the opposite, and bluntly: the announced eleven without Foden is the best we have measured at City this season.
Putting him back would cost points
It runs against the instinct, so here is the working. Rather than asking who replaces Foden — he has no settled position, more on that below — we put him back into the announced eleven in place of each starter in turn, and read the gap.
| If Foden played, instead of… | Effect on win probability |
|---|---|
| Iliman Ndiaye | −0.27 pts |
| Enzo Fernández | −2.12 pts |
| Antoine Semenyo | −2.16 pts |
| Rayan Cherki | −3.88 pts |
Those are the four slots Pep Guardiola and then Enzo Maresca have used him in across the year. In none of them would his return improve the side, and in three it would set it back by two points or more.
It has to be said plainly: it is not that his suspension costs nothing. It is that Sunday's eleven is better without him.
The announced line-ups
1
27
6
3
24
5
17
7
10
42
9
9
10
28
12
27
34
6
5
4
20
22Line-ups as announced before kick-off. The highlighted player is the one filling the vacant position.
What changed around him
Rayan Cherki and Phil Foden are not alternatives to one another: they have started nine matches together in 2026. They are two starters in the same area of the pitch, not two candidates for one place.
That was, in fact, the first call the new manager made. Enzo Maresca succeeded Pep Guardiola on 1 July, after a ten-year reign, and the question waiting for him at Manchester was exactly that one: which of the two? He answered "both".
| Under Maresca, in five matches | Starts | Minutes | Per match |
|---|---|---|---|
| Phil Foden | 4 | 361 | 72 min |
| Rayan Cherki | 4 | 283 | 57 min |
So Foden plays more than Cherki. The model measures what each produces once the quality of team-mates and opponents is stripped out — and on that ground the gap is clear. It is not that the manager is wrong or the model right: it is a disagreement, and it is more interesting than the suspension.
The players City cannot do without
The same calculation, applied to all eleven starters, gives the hierarchy of the day.
| If this player were missing | Replaced by | Cost |
|---|---|---|
| Rayan Cherki | Phil Foden | 3.88 pts |
| Matheus Nunes | Rico Lewis | 1.83 pts |
| Enzo Fernández | Tijjani Reijnders | 1.77 pts |
| Marc Guéhi | Abdukodir Khusanov | 1.32 pts |
| Erling Haaland | Jérémy Doku | 1.27 pts |
| Antoine Semenyo | Ryan McAidoo | 1.20 pts |
| Josko Gvardiol | Abdukodir Khusanov | 0.79 pts |
| Rúben Dias | Abdukodir Khusanov | 0.49 pts |
Stockport, and afterwards
Phil Foden joined City's academy as a child and has been under contract there since 2017. He is twenty-six. In between he was the heir apparent, the boy they protected, then the best player in the Premier League.
He is now one of five attacking midfielders in a squad that has bought four of them in fifteen months — Cherki, Semenyo, Ndiaye, Enzo Fernández — and the only one our model says the team plays better without. On Sunday the Etihad hosts a promoted side with a win probability close to three in four, and nobody in the ground will need to ask who is playing instead of number 47.
That, perhaps, is the news — and it is not good news for him.
See the table and projections for Premier League →
The figures in this piece are read from the latest published computation: they update with the model and may therefore differ from those of the publication date. That is deliberate — an article should not go on asserting what is no longer true.