The model on this site does not read headlines or remember reputations. It watches who played whom and how the game finished, then adjusts. After the first round of matches it moved some teams sharply — and the interesting question is not that it moved, but which of those moves you should actually believe.

How the numbers move

Every result nudges both teams' ratings. The size of the nudge depends on the surprise: beat a side the model rated well above you and you gain a lot; lose to one it rated below you and you drop hard. A 7–1 thumping of a minnow barely shifts a favourite, because that is roughly what was expected. A clear win over a peer moves the needle much more. With one game played, every figure below is built on a single data point — useful, but loud.

Three rises worth trusting

  • Australia (+35) — the biggest climb of the round, and the one I'm most convinced by. Beating Türkiye 2–0 is not a fluke scoreline; a two-goal margin with a clean sheet against a higher-rated side is exactly the kind of result the model is built to reward. This looks like a genuine reassessment, not a bounce.
  • United States (+24) — 4–1 over Paraguay was emphatic, and the attacking output is the part I trust most. Four goals is hard to fake. Home advantage inflates the mood a little, but the margin is real.
  • Ivory Coast (+22) — the quietest of the three and the most underrated. A 1–0 win over Ecuador, a side the model had ahead of them, is the definition of a result that should change your priors. Low-scoring, but earned.

Three falls — real or just one bad night?

  • Türkiye (−35) — the steepest drop. It's tempting to call −35 an overreaction, but I don't think it is. Losing 0–2 to a lower-rated opponent without scoring is a poor night by any reading. The talent is there; the model is simply waiting for it to show up.
  • Paraguay (−24) — conceding four is alarming, but the context matters: this was away to a motivated host. I'd treat this drop as half-real. The defence has a question to answer, not a verdict against it.
  • Ecuador (−22) — the most suspect fall on the board. Losing 0–1 to Ivory Coast is a bad result, not a humiliation, and a single tight defeat can read as noise more than decline. If I'm doubting one of these drops, it's this one.

What to watch next

Two fixtures will tell us whether the model got it right. United States vs Australia pits the round's two biggest risers against each other — a real test of whether both deserved their bump. And Türkiye vs Paraguay throws the two heaviest fallers together: someone has to climb back. Watch those, and you'll know which of these early moves was signal.