Was the forecast right for San Francisco on Fri Sep 4?
San Francisco, California · daily high temperature in °C · forecasts from each model's archived runs, verified against the SFOC1 observations
13 days out, 4 models put San Francisco's Fri Sep 4 high anywhere from 19° to 24°;by the final call, the day before, the field had settled to a median of 20°. The day verified at 20° against the SFOC1 observations — cloudy.
AIFS landed closest, off by 0.2°. GEM ran 2° cool.
What every model said, at every lead
Daily-high calls for Fri Sep 4 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 · 2026-08-22 | 23° | 24° | 19° | — | 23° | — | — | 19–24° |
| 12d · 2026-08-23 | 21° | 30° | 17° | — | 22° | — | — | 17–30° |
| 11d · 2026-08-24 | 23° | 22° | 19° | — | 21° | — | — | 19–23° |
| 10d · 2026-08-25 | 25° | 30° | 20° | — | 21° | — | — | 20–30° |
| 9d · 2026-08-26 | 24° | 21° | 18° | 17° | 22° | — | — | 17–24° |
| 8d · 2026-08-27 | 20° | 26° | 20° | 21° | 20° | — | — | 20–26° |
| 7d · 2026-08-28 | 20° | 21° | 20° | 18° | 20° | — | — | 18–21° |
| 6d · 2026-08-29 | 20° | 23° | 18° | 19° | 22° | 18° | 21° | 18–23° |
| 5d · 2026-08-30 | 20° | 22° | 19° | 19° | 18° | 17° | 22° | 17–22° |
| 4d · 2026-08-31 | 20° | 21° | 19° | 18° | 19° | 17° | 20° | 17–21° |
| 3d · 2026-09-01 | 20° | 21° | 18° | 18° | 20° | 19° | 20° | 18–21° |
| 2d · 2026-09-02 | 20° | 21° | 18° | 18° | 20° | 18° | 22° | 18–22° |
| 1d · 2026-09-03 | 20° | 21° | 19° | 18° | 21° | 18° | 21° | 18–21° |
| actual | 20° |
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 the SFOC1 observations. The actual values are the ones in hand when this receipt was settled, usually 2 days after the date; a correction published after that is not carried into it. Archive depth varies by city; receipts exist only for dates the archive watched arrive.