Scotland — Brazil 0:3: Three Answers, No Reply

Brazil put three past Scotland in their FIFA World Cup group meeting on 24 June 2026, winning 0:3 on the road — and still finished the night as the model's biggest underachiever. An xG of 4.33 is the kind of number a side posts once in a tournament, and converting it into three goals actually left value on the table. The 52% anomaly reading points at the real story: two goalkeepers who refused to follow the script.

An xG of 4.33: how Brazil built the win

The visitors produced 21 attempts, 14 of them inside the box, and put 9 on target. That distribution is what a 4.33 xG looks like in practice — not one golden chance but a steady procession of good ones.

The finishing lagged behind the creation. A conversion of 0.69 means Brazil scored more than a goal less than the model expected, a gap that usually costs points. Here the volume was so large it merely trimmed the margin.

The passing platform matched the pattern: 605 passes to 500, with 564 accurate against 450. Notably, that midfield gap is far narrower than the shooting gap — Scotland stayed in the game everywhere except the two penalty areas.

Scotland's 14 shots and the goal that never came

The hosts were no bystanders. They generated 14 attempts, 10 from inside the box, worth 1.04 xG — a full goal by the model's accounting. Five efforts hit the target.

All five were saved. A blank scoresheet from that output is not an indictment of Scotland's attack so much as a receipt from Brazil's goalkeeper, who finished with 5 saves and every one of them necessary.

Five saves each and +1.23 apiece: the goalkeepers' match

Both keepers ended on exactly 5 saves, and both posted a Goals Prevented figure of +1.23 — each stopped considerably more than the model demanded. Mirror-image excellence like that is rare, and it reshaped the scoreline from both ends.

It is why 0:3 reads stranger than it looks. Brazil's keeper erased Scotland's entire xG; Scotland's keeper clawed back more than a goal of Brazil's mountain of chances and still conceded three, because the mountain was simply too high.

Blocks and misses: the 22 attempts that vanished

Brazil sprayed 7 efforts off target and had another 5 blocked — Scotland's defenders kept throwing themselves into lanes long after the pattern of the match was settled, and that resistance is why a 4.33 xG night ended at three rather than five. Scotland's own wastage was proportionate: 6 off target, 3 blocked.

Those columns matter for reading the scoreline honestly. The hosts lost the quality battle decisively, but they never stopped subtracting from Brazil's total wherever a body could be placed.

Corners 7 to 6 for the losers: the supporting columns

Scotland actually won the corner count, 7 to 6, and held 46% of the ball — hardly a siege pattern. Fouls were level in spirit at 10 to 11, the hosts stayed onside all night while Brazil strayed once, and the cards read 1 yellow to 2.

Every peripheral number says this was a contest. The scoreline says otherwise, and the difference between the two lives entirely in shot quality: Brazil turned similar territory into chances worth four times as much, and that single superiority made all the equal columns irrelevant.

What the match report leaves out

Possession split 46-54 in favour of Brazil — a gap that barely shows up on the scoreline but shapes everything else in the report. On the passing numbers, Scotland finished at 90.0% and Brazil at 93.2% — the side dictating circulation wasn't necessarily the side creating the better chances.

Scotland were penalised 10 times, Brazil 11 times; the report doesn't break fouls down by zone, so it's hard to tie the number directly to chances conceded. The offside count (0-1) is modest, but it tends to move with the number of balls played over the top rather than total attacking volume.

The source for every number here is the official Scotland-Brazil match sheet — none of it is a manual estimate.