The Spread

Which weather model is most accurate in Houston?

Houston, Texas · daily high temperature in °F · trailing 30 days · refreshed at most every 6 hours

Every day, The Spread archives what each major weather model predicted for Houston at 1-, 3-, and 7-day leads, then scores those forecasts against ERA5 reanalysis once the day has happened. This page covers the trailing 30 days (2026-07-12 to 2026-08-11), for daily high temperature.

One day out, ECMWF IFS has been most accurate (typical miss 1.65°F); at three days, ECMWF AIFS (AI) (1.85°F); at seven days, ECMWF AIFS (AI) (2.64°F).

The largest standing bias belongs to NOAA GFS, which has run warm by about 5.36°F at 3-day lead over this window.

The Spread's own bias-corrected blend scored a 1.16°F 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 °F 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 1.6° (-0.5°) 2.1° (-0.0°) 2.6° (+0.4°) 31
DWD ICON 1.7° (+0.8°) 2.5° (+1.7°) 31
NOAA AIGFS (AI) 1.8° (+0.2°) 2.3° (-0.2°) 3.5° (-0.5°) 31
NOAA GFS 1.9° (+0.5°) 5.8° (+5.4°) 6.6° (+6.4°) 31
ECMWF AIFS (AI) 1.9° (-1.4°) 1.9° (-1.1°) 2.6° (-1.6°) 31
GEM (Canada) 2.2° (+1.5°) 2.1° (+0.9°) 2.9° (+1.1°) 31
UKMO (UK) 2.7° (+2.5°) 3.0° (+3.0°) 31
NBM (US blend) 1.4° (-0.6°) 1.8° (+0.1°) 2.9° (+0.5°) 31
Blended (bias-corrected) 1.2° (+0.0°) 1.3° (+0.0°) 2.4° (+0.0°) 31

Method, honestly

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 ERA5 reanalysis daily highs.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.

  • Sun Aug 9 13 days out the models said 90–102; it hit 90. ICON led (off 0.4°F); UKMO ran 6°F hot.
  • Sat Aug 8 12 days out the models said 90–99; it hit 94. GEM led (off 0.5°F); ECMWF ran 5°F cold.
  • Fri Aug 7 11 days out the models said 86–98; it hit 94. ECMWF led (off 0.2°F); GFS ran 4°F hot.
  • Thu Aug 6 10 days out the models said 86–103; it hit 91. ECMWF led (off 1.5°F); AIGFS ran 7°F hot.
  • Wed Aug 5 9 days out the models said 87–103; it hit 96. UKMO led (off 0.0°F).
  • Tue Aug 4 8 days out the models said 89–102; it hit 96. GEM led (off 0.0°F); ICON ran 3°F cold.
  • Mon Aug 3 7 days out the models said 89–101; it hit 99. ECMWF led (off 0.3°F); AIFS ran 4°F cold.
  • Sun Aug 2 6 days out the models said 91–100; it hit 100. UKMO led (off 0.3°F); AIFS ran 6°F cold.
  • Sat Aug 1 5 days out the models said 94–105; it hit 97. GFS led (off 0.1°F); UKMO ran 3°F hot.
  • Fri Jul 31 4 days out the models said 94–102; it hit 98. GEM led (off 0.0°F); AIFS ran 3°F cold.
  • Thu Jul 30 3 days out the models said 93–103; it hit 97. ECMWF led (off 0.5°F); AIFS ran 3°F cold.

Other cities

Atlanta, Georgia Boston, Massachusetts Buffalo, New York Charlotte, North Carolina Chicago, Illinois Dallas, Texas Denver, Colorado London, England Los Angeles, California Miami, Florida Milan, Lombardy Minneapolis, Minnesota New Orleans, Louisiana New York, New York Oklahoma City, Oklahoma Orlando, Florida Phoenix, Arizona Salt Lake City, Utah San Francisco, California Seattle, Washington Washington, District of Columbia

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