Whoever created more clear chances won all six matches. That proves almost nothing.

Six Champions League fixtures were played last night. For each of them we pulled the full statistics sheet — possession, shots, clear chances, passes, dangerous attacks — along with the minute of every goal.

One statistic ranks all six results correctly. It is neither possession nor shot count.

Clear chances, six times out of six

Six times out of six, the side that created more clear chances won. No exception, including in the four fixtures decided by a single goal.

Possession, meanwhile, says nothing

Three of the six winners had the ball less than half the time, two of them barely more than a third. Across these six matches, knowing who dominated possession tells you nothing about who won.

Real Madrid — Inter: having the ball, or having the chances

This is where the gap is most violent.

Inter had 64% possession, completed 582 passes to 305, took 21 shots to 16, won 11 corners to 4, and mounted 94 dangerous attacks to 39. On almost every line of the sheet, Inter overwhelm.

Real Madrid created 9 clear chances to 4, hit the target 8 times to 7 — and won 2–1.

The goals tell the rest: 14th and 23rd minute for Real, 77th for Inter. The match was settled before the half-hour, and the hour of domination that followed produced one goal.

Porto: the best defence of our six leagues, and no shot on target

Four days ago we wrote that Porto have, with an index of 0.56, the best defence across the six leagues we cover, all teams included.

Last night, at home against Manchester City, Porto finished with:

Five shots attempted, none on target, no clear chance, 36% possession. City scored in the 47th minute and again in the 90th.

That is not a refutation of our model, and it matters to understand why.

What the model said about the clubs involved

Nine of the twelve clubs that played last night appear in our six leagues. Here is where they stand at home, which remains the only comparison we can defend.

Real Madrid		1.66
Real Betis		1.23
Villarreal		1.46

One last mention goes to Aston Villa. They are 18^th in the Premier League, with an attack of 1.15 and a defence of 1.10 — to a very close approximation, an average side in their league. They went and won at Brugge with 36% possession and 21 shots, leading 3–1 at half-time.

Here too, one match demonstrates nothing. An average Premier League side regularly wins fixtures it was not supposed to win: that is what a probability means.

What it took before we dared publish a probability

Putting a figure on Real Madrid — Inter meant first solving the Porto problem: how far is each league offset from the others?

We added two terms per country to the Dixon-Coles model — one for attack, one for defence — and estimated them on 705 European fixtures spread over eight seasons. For each one, both clubs' strengths were recomputed as of the week of the match, from the data preceding it alone: no fixture takes part in estimating its own strengths.

Whether those terms earn their place is a separate question. So we tested them on a full season never seen during fitting, 138 fixtures:

The second line is the telling one. Comparing two domestic indices with no correction — what anyone does when they put two figures from this site side by side — barely beats betting at random on historical frequencies. That is exactly what we wrote above, now measured.

What you will see, and what you will not

The Champions League and Europa League pages now show a probability for every fixture whose two clubs both play in one of our six leagues.

That is not many: at the time of writing, three fixtures out of twelve on the next Champions League matchday, two out of sixteen in the Europa League. Everywhere else — Barcelona against Feyenoord, Bayern against Bodø/Glimt — the cell stays empty, because we model neither the Netherlands nor Norway, and inventing the level of an opponent we do not measure would be precisely what we criticise in others.

An empty cell is not an oversight. It is the limit of what we know, written where it falls.