Forecast accuracy log
Every few hours this app writes down what it thinks the weather will be, at nine different distances into the future. Separately, it writes down what the weather actually did. This page scores one against the other — and scores four rival ways of blending the same sources, so the method the app uses has to keep earning it.
One of those rivals is Open-Meteo’s own blend, left completely alone. It already mixes the same global models, so if our re-blending is making things worse rather than better, that line will beat “Consensus (live)” here — and then we switch.
reading the log…
Does the wording match the data?
Every couple of hours the app rebuilds the exact sentences it would show at the top of the page, then checks them against the exact numbers they came from. A failure here never means the weather surprised us — it means the page disagreed with itself, which is always a bug.
reading the audit…
Reported issues
Reports are private. Enter the maintainer key to read and triage them.