Was the forecast right for San Francisco on Sun Aug 9?
San Francisco, California · daily high temperature in °C · forecasts from each model's archived runs, verified against ERA5
13 days out, 4 models put San Francisco's Sun Aug 9 high anywhere from 17° to 24°; by the final call the field had settled to a median of 23°. ERA5 says the day verified at 19° — partly cloudy.
ICON landed closest, off by 1.4°. AIGFS ran 9° hot.
What every model said, at every lead
Daily-high calls for Sun Aug 9 as each model's runs approached it — the last archived run per day. NBM and The Spread's own blend are consensus products and stay out of these ranges and rankings; the San Francisco scoreboard shows them, labelled.
| Lead | AIFS | AIGFS | ECMWF | GEM | GFS | ICON | UKMO | Range |
|---|---|---|---|---|---|---|---|---|
| 13d · 2026-07-27 | 20° | 24° | 17° | — | 22° | — | — | 17–24° |
| 12d · 2026-07-28 | 21° | 19° | 19° | — | 22° | — | — | 19–22° |
| 11d · 2026-07-29 | 21° | 20° | 18° | — | 21° | — | — | 18–21° |
| 10d · 2026-07-30 | 21° | 22° | 18° | — | 20° | — | — | 18–22° |
| 9d · 2026-07-31 | 20° | 22° | 17° | — | 21° | — | — | 17–22° |
| 8d · 2026-08-01 | 20° | 24° | 19° | 20° | 24° | — | — | 19–24° |
| 7d · 2026-08-02 | 21° | 23° | 18° | 20° | 21° | — | — | 18–23° |
| 6d · 2026-08-03 | 21° | 24° | 18° | 19° | 22° | 17° | — | 17–24° |
| 5d · 2026-08-04 | 23° | 26° | 22° | 21° | 23° | 17° | 20° | 17–26° |
| 4d · 2026-08-05 | 23° | 27° | 20° | 22° | 23° | 17° | 19° | 17–27° |
| 3d · 2026-08-06 | 23° | 27° | 22° | 20° | 23° | 16° | 20° | 16–27° |
| 2d · 2026-08-07 | 24° | 27° | 24° | 23° | 24° | 17° | 23° | 17–27° |
| 1d · 2026-08-08 | 23° | 28° | 24° | 24° | 26° | 17° | 22° | 17–28° |
| 0d · 2026-08-09 | 23° | 28° | 23° | 23° | 24° | 17° | 23° | 17–28° |
| actual | 19° |
Read it honestly
This is one date, not a league table. A model can win a day it deserved to lose; rankings only mean something over a window. For that, see which model has been most accurate in San Francisco — or every city The Spread scores.
Forecasts are what each model's archived runs said at the time, captured at The Spread's sweep times — never reconstructed after the fact. Truth is ERA5 reanalysis (temperature_2m_max). Archive depth varies by city; receipts exist only for dates the archive watched arrive.