Which weather model is most accurate in Washington?

Outlook Model accuracy

Washington, District of Columbia · 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 Washington at 1-, 3-, and 7-day leads, then scores those forecasts against the DCA 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 GFS, which has run warm by about 1.98°F at 1-day lead over this window.

The Spread's own bias-corrected blend scored a 1.39°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
UKMO (UK)	1.4° (+0.3°)	2.8° (+1.2°)	—	30
GEM (Canada)	1.7° (-0.5°)	3.6° (-2.0°)	3.5° (-0.8°)	30
NOAA AIGFS (AI)	1.9° (-0.8°)	2.3° (-1.0°)	3.5° (-1.2°)	30
DWD ICON	2.0° (-0.9°)	2.7° (+0.7°)	—	30
ECMWF AIFS (AI)	2.0° (-1.4°)	2.1° (-1.0°)	3.2° (-0.7°)	30
ECMWF IFS	2.4° (+0.2°)	3.4° (-0.0°)	4.1° (+0.1°)	30
NOAA GFS	2.6° (+2.0°)	3.3° (+1.8°)	4.7° (+3.2°)	30
NBM (US blend)	1.9° (-1.1°)	2.7° (-0.9°)	2.8° (-1.7°)	30
Blended (bias-corrected)	1.4° (-0.0°)	2.3° (+0.0°)	2.5° (-0.0°)	30

Method, honestly

The Spread's Washington point is the Washington National station (DCA), 3.7 mi 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 DCA 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 12 days out the models said 75–98; it hit 98. GFS led (off 0.9°F); AIGFS ran 5°F cold.
Mon Aug 31 11 days out the models said 75–94; it hit 87. ICON led (off 0.8°F); GFS ran 7°F hot.
Sun Aug 30 10 days out the models said 75–96; it hit 89. UKMO led (off 0.2°F); GFS ran 5°F hot.
Sat Aug 29 9 days out the models said 82–91; it hit 85. ICON led (off 0.2°F); AIGFS ran 3°F cold.
Fri Aug 28 8 days out the models said 85–90; it hit 90. GFS led (off 0.1°F); AIFS ran 5°F cold.
Thu Aug 27 7 days out the models said 79–87; it hit 88. GEM led (off 0.0°F); GFS ran 4°F hot.
Wed Aug 26 6 days out the models said 82–88; it hit 86. ICON led (off 0.1°F); GFS ran 3°F hot.
Tue Aug 25 5 days out the models said 79–86; it hit 85. GEM led (off 0.9°F); UKMO ran 2°F cold.
Mon Aug 24 4 days out the models said 77–84; it hit 85. AIGFS led (off 1.6°F); ECMWF ran 4°F cold.
Sun Aug 23 3 days out the models said 81–86; it hit 87. GFS led (off 1.0°F); ECMWF ran 4°F cold.
Mon Aug 17 13 days out the models said 80–87; it hit 90. ICON led (off 0.1°F); GEM ran 4°F hot.
Sun Aug 16 13 days out the models said 81–90; it hit 81. UKMO led (off 1.0°F); AIFS ran 4°F cold.
Sat Aug 15 13 days out the models said 83–101; it hit 89. AIFS led (off 0.4°F); GEM ran 3°F hot.
Fri Aug 14 13 days out the models said 75–92; it hit 79. ICON led (off 3.7°F); UKMO ran 10°F hot.
Thu Aug 13 13 days out the models said 84–100; it hit 89. UKMO led (off 0.1°F); GEM ran 6°F hot.
Wed Aug 12 13 days out the models said 78–91; it hit 87. ICON led (off 0.6°F); GEM ran 6°F hot.
Tue Aug 11 13 days out the models said 87–90; it hit 88. AIGFS led (off 0.6°F); GEM ran 4°F hot.
Mon Aug 10 13 days out the models said 90–105; it hit 90. ICON led (off 3.5°F); GEM ran 9°F hot.
Sun Aug 9 13 days out the models said 80–92; it hit 92. ICON led (off 1.2°F); GEM ran 5°F hot.
Sat Aug 8 12 days out the models said 83–94; it hit 90. UKMO led (off 1.4°F); GFS ran 6°F hot.
Fri Aug 7 11 days out the models said 89–100; it hit 90. ICON led (off 1.6°F); GEM ran 6°F hot.
Thu Aug 6 10 days out the models said 88–98; it hit 92. ICON led (off 0.4°F); GEM ran 4°F hot.
Wed Aug 5 9 days out the models said 84–96; it hit 86. ICON led (off 0.1°F); GEM ran 3°F hot.
Tue Aug 4 8 days out the models said 85–90; it hit 88. ICON led (off 0.1°F); GEM ran 3°F hot.
Mon Aug 3 7 days out the models said 83–90; it hit 88. AIFS led (off 0.2°F); AIGFS ran 3°F cold.
Sun Aug 2 6 days out the models said 78–90; it hit 82. ICON led (off 1.2°F); GFS ran 7°F hot.
Sat Aug 1 5 days out the models said 85–91; it hit 90. ECMWF led (off 0.1°F); ICON ran 3°F cold.
Fri Jul 31 4 days out the models said 84–88; it hit 90. GFS led (off 0.2°F); GEM ran 4°F hot.
Thu Jul 30 3 days out the models said 80–86; it hit 87. GFS led (off 0.1°F); GEM ran 4°F hot.

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 Los Angeles, California 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

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