Side-by-side
Compare stations
Up to four airfields · accuracy, win-rates and current forecasts in one shareable view
Station selection
Pick up to four stations — the URL updates so you can bookmark or share the exact comparison.
Daily MAE
Mean absolute error of the published forecast vs. the settled daily maximum, per day.
Checkpoint win-rate matrix
Share of days each station's prediction landed within 1°C of the settled maximum, per checkpoint.
Track record
Rolling window quality metrics from each station's scored editions.
Who was right: engine, prior, or analog?
Every forecast blends three voices — the published engine centre, the external prior it trusted (TAF TX or NWP), and the station's own analog climatology. Each row scores one voice's long-lead forecasts against the settled maximum, so you can see who has been carrying the accuracy and who has been dragging.