Sunderland vs Chelsea
Our model puts a Chelsea win at 49.3% in this Premier League fixture — a moderate lean. Expected goals run 1.12 to 1.64.
What this probability has been worth
A number is worth what it has been worth. Across 2,561 predictions published before kick-off and scored against the result afterwards, the calls we priced between 45% and 55% came in 48.5% of the time — 779 fixtures in that band. The model claimed 49.5%, so the two are within 1 points of each other. In this range the stated probability has meant roughly what it says.
The ratings gap
Chelsea carry a rating of 1796 against 1700 for Sunderland, a gap of 96 points. On our scale that is the difference between a side that controls most of its fixtures and one that has to earn each of them. It is also the single largest input into the probabilities above.
Attack against defence
The single rating hides where the mismatch actually sits. Splitting it, Chelsea's attack should find room against Sunderland's defence, by 103 points. That is the mechanism behind the expected-goals line rather than a separate opinion about it.
Chelsea are off their season peak
Chelsea peaked at 1869 this season and sit at 1796 today, 73 points below their own best. Measuring a side against itself rather than against the table is often the earlier signal, and it is not visible in a league position.
Finding the target
Chelsea struggles to hit the target: 28.7% of their shots are on goal over their last matches. The average across the sides we have figures for is 34%. Accuracy is not the same as scoring — it says how much of a side's shooting is worth a goalkeeper's attention, which is the part that does not swing as wildly from week to week as goals do.
Behind the defence
51% of the shots on target Sunderland face do not end up in the net, against 67.3% for the sides we measure. That is low enough to matter, and it is the kind of number that flatters or punishes a side's record without saying much about how it plays.