United States
Which forecast is actually the most accurate?
One record, the same climate day, and no one pays for a place on the list.
Forecast Scoreboard compares the National Weather Service and weather models on the same climate day. A meteorologist is scored only after they opt in. Television stations are not graded. Ranks wait until the sample is long enough to mean something.
HRRR is the best-ranked model on the next-day board. HRRR has a published rank in 206 markets. Those ranks average about 222 scored days. That board ranks tomorrow's high, low, and rain chance together against the National Weather Service. The National Weather Service baseline still did better on those same days. HRRR's average miss on highs is 2.6°F. Compared with the National Weather Service, which is the baseline and is not ranked against itself.
Next day board Full week board Markets World leaderboard Models 2026 hurricane model accuracy
- Markets
- 210
- Stations listed
- 845
- Climate days on file
- 46,868
- NWS forecasts archived
- 81,742
Find your city
Type a city, a state, or a ZIP code. The result links to that market, scored at one official airport.
Today's risk
Storm Prediction Center severe outlook and Weather Prediction Center excessive rainfall outlook. The lines are the stored official products.
Checked 11:35 PM ET Wed, Nov 11.
Severe storms
Day 1
Issued 0640 PM CST Wed Nov 11 2026. Valid 120100Z - 121200Z.
THERE IS A SLIGHT RISK OF SEVERE THUNDERSTORMS THROUGH MID EVENING ACROSS PARTS OF WESTERN INTO CENTRAL KENTUCKY
Strong thunderstorms posing a risk for small to marginally severe hail, and perhaps locally gusty winds, may be maintained across parts of west central into southeastern Kentucky and adjacent northern Tennessee into at least late evening.
Highest category: Slight.
- Bowling Green, KY · Marginal
- Lexington, KY · Marginal
- Louisville, KY · Marginal
Day 2
Issued 1130 AM CST Wed Nov 11 2026. Valid 121200Z - 131200Z.
THERE IS A MARGINAL RISK OF SEVERE THUNDERSTORMS FOR PORTIONS OF THE MID ATLANTIC
A couple of damaging gusts or marginally severe hail may accompany one of the stronger storms across the Mid-Atlantic tomorrow (Thursday) afternoon.
Highest category: Marginal.
- New York, NY · Marginal
- Philadelphia, PA · Marginal
Day 3
Issued 0123 PM CST Wed Nov 11 2026. Valid 131200Z - 141200Z.
No severe areas forecast.
Severe thunderstorms are not expected Friday into early Saturday morning.
Excessive rainfall
Day 1
Issued 757 PM EST Wed Nov 11 2026. Valid 01Z Thu Nov 12 2026 - 12Z Thu Nov 12 2026.
No excessive rainfall area is forecast.
Day 2
Issued 757 PM EST Wed Nov 11 2026. Valid 12Z Thu Nov 12 2026 - 12Z Fri Nov 13 2026.
No excessive rainfall area is forecast.
Day 3
Issued 757 PM EST Wed Nov 11 2026. Valid 12Z Fri Nov 13 2026 - 12Z Sat Nov 14 2026.
No excessive rainfall area is forecast.
Snow and ice
No snow or ice event is on the calendar for today or the next two days.
Storm Prediction Center convective outlooks · Weather Prediction Center excessive rainfall outlook
Live now
Events that have started and are still inside their dates. An event whose end date is today stays here until that Eastern day ends. A storm also stays while the National Hurricane Center is still issuing advisories for it. A remnant low, a post-tropical cyclone, or a storm that has dropped off that list moves to Just finished. The line under each name is read from the stored record.
Just finished
Events whose end date was before today, Eastern time, within the past week. A storm also moves here on the day of its final advisory. The line is the last stored result.
- Ohio Valley severe storms Nov 11, 2026 60 hail reports, 2 wind reports, and 4 wind-damage reports are on file. Scorecard Damage map
- South Texas flash flood Nov 7-9, 2026 10 flash flood warnings and 20 local storm reports are on file. Scorecard Damage map
- South Texas severe storms Nov 8, 2026 No tornado, hail, or wind report verified for this storm. Scorecard Damage map
- Carolinas and southern Mid-Atlantic severe storms Nov 6, 2026 No tornado, hail, or wind report verified for this storm. Scorecard Damage map
- East Coast and southeastern New England heavy rain Nov 6, 2026 4 flash flood warnings and 44 local storm reports are on file. Scorecard Damage map
- East Florida coast flash flood Nov 2-4, 2026 22 flash flood warnings and 17 local storm reports are on file. Scorecard Damage map
- Mid-South flash flood Nov 3-4, 2026 29 local storm reports are on file. Scorecard Damage map
- Southeast and Tennessee Valley severe storms Nov 4, 2026 2 wind reports and 5 wind-damage reports are on file. Scorecard Damage map
New storm pages
Storm and event pages, newest first. RSS feed.
- Ohio Valley severe storms Ohio Valley severe storms, November 11, 2026.
- South Texas severe storms South Texas severe storms, November 8, 2026.
- South Texas flash flood South Texas flash flood, November 7, 2026 to November 9, 2026.
- East Coast and southeastern New England heavy rain East Coast and southeastern New England heavy rain, November 6, 2026.
- Carolinas and southern Mid-Atlantic severe storms Carolinas and southern Mid-Atlantic severe storms, November 6, 2026.
- Southeast and Tennessee Valley severe storms Southeast and Tennessee Valley severe storms, November 4, 2026.
- Mid-South flash flood Mid-South flash flood, November 3, 2026 to November 4, 2026.
- Eastern Kansas, western Missouri, and eastern Oklahoma heavy rain Eastern Kansas, western Missouri, and eastern Oklahoma heavy rain, November 2, 2026 to November 3, 2026.
How to read this. A shorter miss is closer to the weather that happened. Beats the NWS means a smaller miss than the National Weather Service on the same days. Behind the NWS means a larger miss. Early read means at least 90 days scored. Proven means 180.
The technical details
Skill is 1 minus this forecast’s mean loss divided by the National Weather Service mean loss on the same days. Loss is the mean absolute error for temperature and the Brier score for rain chance. Zero is a tie. Positive beats the Weather Service. Negative falls behind it.
Days scored are matched days: issue dates with a forecast from this source and from the National Weather Service, and a final observation. A rank starts at 90 matched days. That status is provisional, shown here as Early read. At 180 matched days the status is established, shown here as Proven. Below the first line the row is still collecting and is not a rank.
How sure is a 95% confidence interval: a percentile bootstrap of the score, resampling the matched days 2,000 times. Overlapping intervals are marked tied.
The headline score uses days 1 through 5, with lead weights 0.30, 0.25, 0.20, 0.15, and 0.10. Days 6 and 7 stay out of it. Variable weights are high 0.35, low 0.30, and rain chance 0.35. Observations come from ACIS, with the National Weather Service climate report when ACIS has a hole. Model numbers come from each provider’s public files.
The methodology page is the full reference. How we grade says the same rules in plain English.
Model leaderboards
Compared with the National Weather Service on the same days. The Weather Service is the baseline, so it is not ranked against itself. A model is listed after 90 days scored in at least 30 markets. That rank is an early read until 180 days scored, then it is proven. NAM and HRRR cover about one or two days, so they are on Next day and are not on Full week.
Includes archived forecasts back to March 31, 2026.
Next day
Tomorrow only. Every model that publishes a next-day high, low, and rain chance can rank here, including the short-range models.
- 1 HRRR Behind the NWS by 17% 2.6° high error
- 2 ECMWF IFS Behind the NWS by 45% 2.9° high error
- 3 NAM Behind the NWS by 46% 3.0° high error
| Place on this list. 1 was closest to the weather that happened. A tie shares this number, and the next place skips ahead. Early read until the longer day count, then proven. | The forecast source this row scores. | Mean skillBeats the National Weather Service, or falls behind it, averaged across markets with a published rank. 0 is the same, positive is better, negative is worse. The technical name is mean skill. | High errorAverage miss on daily highs, in degrees F. Lower is better. | How many markets this model has a published rank in. A national row needs at least 30. | Matched daysDays scored. How many climate days this number is averaged over. The technical name is matched days. | StatusEarly read means at least 90 days scored. Proven means 180 or more. Still collecting means it is too soon to rank. The technical names are provisional and established. |
|---|---|---|---|---|---|---|
| 1 | HRRR | Behind the NWS by 17% | 206 | 222 | Proven | |
| 2 | ECMWF IFS | Behind the NWS by 45% | 210 | 222 | Proven | |
| 3 | NAM | Behind the NWS by 46% | 207 | 215 | Proven | |
| 4 | GFS | Behind the NWS by 46% | 210 | 223 | Proven | |
| 5 | ECMWF AIFS | Behind the NWS by 67% | 210 | 222 | Proven |
The bar is the average miss on highs. A shorter bar is closer to the high that happened. Beats the NWS means a smaller miss than the National Weather Service on the same days.
Full week
Leads 1 through 7, equal weight on each lead that both this model and the National Weather Service forecast, then the same high, low, and precipitation weights as the methodology. Only sources that cover the whole week are eligible. NAM and HRRR are not, and they are not given a full-week number.
No full-week model has cleared the line in 30 markets yet.
Markets
Every US television market is on the scoreboard, each checked at one official airport climate site. The list is not an accuracy order, and the stations in a market are not graded.
Browse 210 markets by state · National Weather Service forecast accuracy
Raw models
Models are forecast sources. They are scored on the same climate day against the National Weather Service. A feed that is not public is listed and left empty.
| The weather model. Models are scored on their own against the National Weather Service. | Whether this model is being collected. Not publicly available means it is listed and left empty, and no number is guessed. |
|---|---|
| Canadian GEM (GDPS) | Collecting from ECCC/MSC Datamart |
| ECMWF AIFS | Collecting from ECMWF open data |
| ECMWF IFS | Collecting from ECMWF open data |
| GEML | Public data, collector not wired |
| GFS | Collecting from NOAA Open Data |
| Google WeatherNext / GraphCast output | Not publicly available |
| HRRR | Collecting from NOAA Open Data |
| ICON | Collecting from DWD Open Data |
| NAM | Collecting from NOAA Open Data |
| UKMET | Not publicly available |
How a number gets onto this site
-
Capture
Each poll is archived with a timestamp and a hash. The forecast of record is the last confirmed version at or before 23:00 local time. Day 1 is the next calendar day.
-
Review
Numbers read off a graphic wait in a review queue. They do not count until a person confirms or corrects them. The NWS feed is structured, so those values are confirmed automatically.
-
Verify
The scored weather is the official climate day: midnight to midnight, local standard time, from ACIS, with the NWS daily climate report as a fallback. A same-day “as of” report is stored and not scored.
Get the storm report card
Leave an email address and the storm report card can go there when it is ready.
We store this address for a future storm report card. We do not sell it. No email is sent yet.