Liverpool 0-3 Manchester City: what the three goals at Anfield are worth in the title race
Before kick-off, our model gave City 42.6%. The score was level when Gravenberch was sent off, then City scored three times. We replayed the season with a different score: the gap is worth 15.4 points of title probability.
By Lucas da Silva · 12 October 2026 · 5 min read · Models by Rondo Research
The score was level when Ryan Gravenberch was shown his second yellow card. It was 0-3
afterwards. In between, a game that could still be read as balanced turned into a
demonstration by 
Manchester City, who stay on 6 wins from 6 games.
The debate over the sending-off will go on. What interests us is colder: what are three goals scored against ten men worth in the title race?
Before kick-off, the game was open
Our model, calculated on the morning of 11 October, gave City a 42.6% chance of winning,
Liverpool 30.5% and the draw 26.9%. The exact score of 0-3 carried only 3.6%. Nobody should
have expected a one-sided game, and for an hour there wasn’t one.
We replayed the season with a different score
The calculation published on 12 October gives City a 67.5% chance of the title. To measure what the Anfield score weighed, we replaced the 0-3 with other outcomes, refitted the strengths of all twenty clubs and replayed the season 50,000 times, exactly as the published calculation does — strengths refitted 60 times. With the real score, we land back on its figures.
A 0-0 at Anfield would have left Arsenal on a 35.4% title chance; with the real score it falls to 24.7%. And a 1-0 Liverpool win would have made the race almost even: 43.6% for City, 39.7% for Arsenal. A single game in October does not decide a title, but this one moved the race by a measurable amount.
| Manchester City | 67.5% |
 Arsenal | 24.7% |
 Brighton & Hove Albion | 6.5% |
Liverpool pay twice
For Liverpool, the cost is measured not in the title but in the Champions League. With the 0-3, the club have a 26.4% chance of a top-4 finish. A 0-0 would have left them 41.3%: the three goals cost them 14.9 points.
There is a less gloomy reading, though. Liverpool are 9th with 9 points, but the model ranks them 4th by strength. This is not a team playing badly: it is a team whose results, six games in, lag behind its level, and which has just lost with ten men to the best club in the league.
What would prove the model wrong
The model reads goals, not how they come. It does not know that part of this game was played ten against eleven, and it treats this 0-3 like any other: it draws from it a City slightly stronger and a Liverpool slightly weaker than they may really be. That is also why we publish the 0-0 alongside it: the truth about the two teams’ strength probably lies in between.
City are 3 points clear of Arsenal, second on 15 points. Calculation of 12 October 2026: at Anfield, City did not just win a game, they gained 15.4 points of title probability after the sending-off.
Frequently asked questions
What are Manchester City’s title chances after their win at Anfield?
The 12 October 2026 calculation gives Manchester City 67.5%, Arsenal 24.7% and Brighton 6.5%.
What would have happened if the game had stayed 0-0?
Replaying the season with a 0-0, City’s title probability drops to 52.1%, against 67.5% with the real score. Liverpool would have had a 41.3% chance of a top-4 finish, against 26.4%.
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.
