The Spread

Which weather model is most accurate in Los Angeles?

Outlook Model accuracy

Los Angeles, California · daily high temperature in °C · trailing 30 days · refreshed at most every 6 hours

Every day, The Spread archives what each major weather model predicted for Los Angeles at 1-, 3-, and 7-day leads, then scores those forecasts against the FHMC1 observations once the day has happened. This page covers the trailing 30 days (2026-08-04 to 2026-09-02), for daily high temperature.

One day out, too close to call; at three days, too close to call; at seven days, too close to call.

The largest standing bias belongs to NOAA AIGFS (AI), which has run cool by about 5.56°C at 3-day lead over this window.

The Spread's own bias-corrected blend scored a 0.57°C typical miss at one day. Its corrections come from the same window it is scored on (in-sample), so read it as a ceiling, not a promise.

The scoreboard

Typical miss (MAE) in °C with the signed bias in parentheses — positive bias means the model ran warm. Days = scored days behind the 1-day number.

Model 1-day 3-day 7-day Days
ECMWF IFS 0.8° (-0.1°) 0.9° (+0.1°) 2.2° (-1.4°) 30
NOAA GFS 0.9° (-0.5°) 1.0° (+0.0°) 1.4° (-0.5°) 30
UKMO (UK) 1.0° (-0.9°) 1.4° (-1.3°) 30
GEM (Canada) 1.1° (+1.0°) 1.0° (-0.2°) 1.4° (+0.0°) 30
DWD ICON 1.8° (+1.7°) 2.4° (+2.4°) 30
ECMWF AIFS (AI) 2.9° (-2.9°) 2.5° (-2.4°) 2.8° (-2.7°) 30
NOAA AIGFS (AI) 5.2° (-5.2°) 5.6° (-5.6°) 5.7° (-5.7°) 30
NBM (US blend) 2.6° (-2.6°) 2.9° (-2.9°) 3.5° (-3.4°) 30
Blended (bias-corrected) 0.6° (-0.0°) 0.8° (+0.0°) 1.3° (+0.0°) 30

Method, honestly

The Spread's Los Angeles point is the Los Angeles Downtown station (FHMC1), 2.2 km from downtown, used for both forecasts and verifications. Forecasts come from each model's actual archived runs via Open-Meteo's Previous Runs API — what the model said at the time, not a reconstruction. Truth is FHMC1 observations — the station's own daily highs, a thermometer rather than a grid, and no correction applies: the thermometer is the truth outright. Rankings cover the models themselves; the two consensus products at the bottom of the table — NBM, which is itself a blend of models, and The Spread's own bias-corrected blend — are shown for comparison, never ranked or named "most accurate" here. The blend's corrections come from the same trailing window it is scored on (in-sample), so treat its score as a ceiling. A rolling 30-day window means these numbers move — that is the point.

Receipts — was the forecast right?

One page per settled date: what every model said as the day approached, what happened, and who was closest.

  • Tue Sep 1 9 days out the models said 28–33; it hit 29. GFS led (off 0.0°C); AIGFS ran 4°C cold.
  • Mon Aug 31 8 days out the models said 26–34; it hit 30. GFS led (off 0.6°C); AIGFS ran 4°C cold.
  • Sun Aug 30 7 days out the models said 28–36; it hit 33. GFS led (off 0.2°C); AIGFS ran 5°C cold.
  • Sat Aug 29 6 days out the models said 29–40; it hit 36. ECMWF led (off 0.1°C); AIGFS ran 5°C cold.
  • Fri Aug 28 5 days out the models said 29–42; it hit 36. UKMO led (off 0.1°C); AIGFS ran 5°C cold.
  • Thu Aug 27 4 days out the models said 28–40; it hit 34. UKMO led (off 0.4°C); AIGFS ran 5°C cold.
  • Wed Aug 26 3 days out the models said 29–41; it hit 37. UKMO led (off 0.2°C); AIGFS ran 5°C cold.
  • Thu Aug 20 13 days out the models said 26–29; it hit 35. GFS led (off 0.9°C); AIGFS ran 8°C cold.
  • Wed Aug 19 14 days out the models said 24–30; it hit 36. ECMWF led (off 0.7°C); AIGFS ran 8°C cold.
  • Tue Aug 18 13 days out the models said 24–32; it hit 35. GFS led (off 0.7°C); AIGFS ran 7°C cold.
  • Mon Aug 17 13 days out the models said 26–32; it hit 34. ECMWF led (off 1.8°C); AIGFS ran 6°C cold.
  • Sun Aug 16 13 days out the models said 25–32; it hit 32. GFS led (off 2.0°C); AIGFS ran 7°C cold.
  • Sat Aug 15 13 days out the models said 25–32; it hit 30. ECMWF led (off 1.7°C); AIGFS ran 5°C cold.
  • Fri Aug 14 13 days out the models said 25–32; it hit 29. ECMWF led (off 1.2°C); AIGFS ran 4°C cold.
  • Thu Aug 13 13 days out the models said 24–30; it hit 30. GFS led (off 1.1°C); AIGFS ran 6°C cold.
  • Wed Aug 12 13 days out the models said 25–33; it hit 32. GFS led (off 1.7°C); AIGFS ran 7°C cold.
  • Tue Aug 11 13 days out the models said 25–31; it hit 33. GFS led (off 2.6°C); AIGFS ran 6°C cold.
  • Mon Aug 10 13 days out the models said 22–31; it hit 34. ECMWF led (off 1.5°C); AIGFS ran 7°C cold.
  • Sun Aug 9 13 days out the models said 27–31; it hit 35. GFS led (off 0.8°C); AIGFS ran 8°C cold.
  • Sat Aug 8 12 days out the models said 27–32; it hit 35. GFS led (off 1.7°C); AIGFS ran 9°C cold.
  • Fri Aug 7 11 days out the models said 28–34; it hit 35. GFS led (off 0.9°C); AIGFS ran 9°C cold.
  • Thu Aug 6 10 days out the models said 28–35; it hit 33. ECMWF led (off 0.0°C); AIGFS ran 7°C cold.
  • Wed Aug 5 9 days out the models said 28–32; it hit 31. ECMWF led (off 0.2°C); AIGFS ran 6°C cold.
  • Tue Aug 4 8 days out the models said 28–33; it hit 31. GFS led (off 0.2°C); AIGFS ran 6°C cold.
  • Mon Aug 3 7 days out the models said 28–35; it hit 33. UKMO led (off 1.0°C); AIGFS ran 7°C cold.
  • Sun Aug 2 6 days out the models said 30–35; it hit 34. ECMWF led (off 0.1°C); AIGFS ran 8°C cold.
  • Sat Aug 1 5 days out the models said 31–36; it hit 35. GFS led (off 0.2°C); AIGFS ran 9°C cold.
  • Fri Jul 31 4 days out the models said 29–35; it hit 35. ECMWF led (off 0.6°C); AIGFS ran 9°C cold.
  • Thu Jul 30 3 days out the models said 29–33; it hit 33. ECMWF led (off 1.0°C); AIGFS ran 8°C cold.

Other cities

Atlanta, Georgia Berlin, State of Berlin Boston, Massachusetts Buffalo, New York Charlotte, North Carolina Chicago, Illinois Dallas, Texas Denver, Colorado Detroit, Michigan Houston, Texas London, England Miami, Florida Milan, Lombardy Minneapolis, Minnesota New Orleans, Louisiana New York, New York Oklahoma City, Oklahoma Orlando, Florida Paris, Île-de-France Region Philadelphia, Pennsylvania Phoenix, Arizona Salt Lake City, Utah San Antonio, Texas San Diego, California San Francisco, California Seattle, Washington St. Louis, Missouri Tampa, Florida Washington, District of Columbia

The Spread — the forecast spread — is in closed beta. These verification pages are public while the rest is gated.