The Spread

Was the forecast right for San Francisco on Wed Aug 19?

San Francisco, California · daily high temperature in °F · forecasts from each model's archived runs, verified against the ECMWF IFS analysis

13 days out, 4 models put San Francisco's Wed Aug 19 high anywhere from 64° to 69°; by the final call the field had settled to a median of 67°. The day verified at 64° against the ECMWF IFS analysis — cloudy.

ICON landed closest, off by 1.5°. AIGFS ran 11° hot.

13 days out the models said 64–69; it hit 64. ICON led (off 1.5°F); AIGFS ran 11°F hot.
The shareable card for this receipt — what this page looks like pasted into a thread.

What every model said, at every lead

Daily-high calls for Wed Aug 19 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 AIFSAIGFSECMWFGEMGFSICONUKMO Range
15d · 2026-08-04 65° 73° 65–73°
14d · 2026-08-05 70° 67° 73° 67–73°
13d · 2026-08-06 64° 69° 66° 68° 64–69°
12d · 2026-08-07 68° 69° 66° 68° 66–69°
11d · 2026-08-08 71° 70° 65° 72° 65–72°
10d · 2026-08-09 67° 84° 70° 74° 67–84°
9d · 2026-08-10 66° 74° 65° 78° 65–78°
8d · 2026-08-11 70° 69° 65° 63° 70° 63–70°
7d · 2026-08-12 70° 70° 66° 62° 70° 62–70°
6d · 2026-08-13 69° 70° 65° 64° 68° 63° 63–70°
5d · 2026-08-14 69° 68° 66° 64° 68° 62° 73° 62–73°
4d · 2026-08-15 71° 73° 69° 65° 72° 64° 71° 64–73°
3d · 2026-08-16 71° 72° 66° 65° 71° 64° 69° 64–72°
2d · 2026-08-17 73° 74° 66° 67° 73° 64° 68° 64–74°
1d · 2026-08-18 70° 74° 66° 65° 69° 64° 65° 64–74°
0d · 2026-08-19 70° 75° 66° 67° 73° 63° 66° 63–75°
actual 64°

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 ECMWF IFS analysis (temperature_2m_max). Archive depth varies by city; receipts exist only for dates the archive watched arrive.