Forecast Scoreboard Forecast
Scoreboard

Days 1–5. Compared with the National Weather Service forecast from the same hour. No one pays for a rank.

Hurricane tracks

Tropical Storm Odalys Model Accuracy: DeepMind vs GFS vs WeatherNext 3

Final advisory

NHC issued its final advisory at 2:00 PM ET Sun, Sep 27.

Tropical Storm Odalys has 45 positions on the best track, which is where the storm actually went, from the operational NHC file. 4 official forecast tracks are drawn. At 48 hours, the average track miss is 44 miles across 23 forecasts. At 48 hours, the official wind forecast averaged 77 mph and where the storm actually went averaged 89 mph. Stored model tracks on these forecast tracks: GFS, GEFS mean, HAFS-A, HAFS-B, HWRF, HMON, NAVGEM, Google DeepMind, WeatherNext 3.

EP162026 · 2026 · Sep 16-27, 2026. Score table. Damage reports.

How to read this. The solid line is where the storm went. A dashed line is where the official forecast said it would go. A smaller path miss is closer. Distances are statute miles. Wind speeds are miles per hour.

The technical details

Path miss on this page is the great-circle distance in statute miles, the same miles as the home card and the score table. The Earth radius is 3,958.7613 miles. The file also keeps that distance in nautical miles. One nautical mile is 1,852 meters, about 1.15 statute miles. A case counts when that forecast and the best track (BEST, lead 0) share the same valid time. Positions are not interpolated. A model track counts even when that cycle has no official forecast yet. Skill against the official forecast (OFCL) still requires the official forecast from the same issue time and lead.

Shown leads are 12, 24, 36, 48, 60, 72, 96, and 120 hours, once a paired fix exists. Sample size is the number of paired cases at that lead. Overall rank still uses 24, 48, 72, and 120 hours. The 96-hour lead is drawn when the deck has a paired fix, and it is not in that rank.

Wind on this page is miles per hour. The file stores maximum wind in knots. One knot is about 1.15 miles per hour. An advisory wind is rounded to the nearest 5 mph. A mean on this page is rounded to the nearest mile per hour. Wind miss is the absolute difference on the paired cases. There is no confidence interval on this map. The daily board’s interval is a 95% percentile bootstrap.

Overall rank is the average of the path rank and the wind rank on this storm. Each rank uses the graded leads at 24, 48, 72, and 120 hours, and each of those leads counts once, however many forecasts it holds. A lower miss is the better rank. A tie shares the average of the places those forecasts occupy. A source is ranked only with at least 3 graded forecasts for path and 3 for wind. Fewer than that says Not enough forecasts yet. A source with no graded path, or no graded wind, is left out of both ranks, and overall shows —.

The best track is the operational NHC b-deck. It is preliminary and revised after the season. HURDAT2 is not loaded. Model lines come from the public a-deck for the same forecast tracks, and only when that file stores them. Google DeepMind is the Weather Lab operational ensemble mean. WeatherNext 3 is that model’s ensemble mean from the same download. ECMWF AIFS, ECMWF HRES, and the ECMWF ENS mean are tropical-cyclone tracks from ECMWF open data. The ECMWF ENS mean is the average of the member centers in the IFS ensemble track file. AI GFS is NOAA’s AIGFS tracker when that file is still posted. The methodology page names the aids that are scored.

The methodology page is the full reference. How we grade says the same rules in plain English.

The best track is the operational NHC b-deck. It is preliminary and is revised after the season. The post-season HURDAT2 track is not loaded.

ECMWF AIFS, ECMWF HRES, ECMWF ENS mean, AI GFS: No graded forecasts yet.

Overall rank

Overall averages each forecast's rank on path and on wind strength. See the separate scores below.

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. Path and wind ranksOverall averages each forecast's rank on path and on wind strength. See the separate scores below. Track rankPlace on path miss. 1 was closest. On one storm this is that place. Across storms it is the average of those places. Each graded lead counts once. A tie shares the average place. Intensity rankPlace on wind miss. 1 was closest. On one storm this is that place. Across storms it is the average of those places. Each graded lead counts once. A tie shares the average place.
1 Google DeepMind 1 1 1
2 NHC official 2.5 3 2
2 WeatherNext 3 2.5 2 3
4 HMON 6.5 5 8
5 HAFS-B 7 7 7
6 HAFS-A 7.5 10 5
7 GEFS mean 8 6 10
7 Canadian GEM 8 4 12
7 NAVGEM 8 12 4
10 GFS 8.5 11 6
10 HWRF 8.5 8 9
12 UKMET 10 9 11

How we grade

Track and intensity by model

Each row is one source at one lead that has already verified, including 12, 24, 36, and 48 hours. The count is how many forecasts shared a time with where the storm went. Path miss is statute miles. Wind miss is miles per hour. The rank number is that source's place in the table above, and it still uses 24, 48, 72, and 120 hours.

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. LeadHow many hours ahead this forecast was made. Scored leads are 24, 48, 72, and 120. The technical name is lead. CasesHow many past forecasts went into this average. A forecast counts only when the best track has a fix at that exact time. The technical name is cases. Track errorAverage how far the predicted storm path was from where it actually went, in miles. Lower is better. The technical name is track error. Track skillHow that path miss compares with the official hurricane forecast. Positive means better than the official forecast. Negative means worse. The technical name is track skill. Intensity errorAverage how far off the predicted wind speed was, in miles per hour. The file stores knots. Lower is better. The technical name is intensity error. Intensity skillHow that wind miss compares with the official hurricane forecast. Positive means better. Negative means worse. The technical name is intensity skill.
1 Google DeepMind 12 h 29 18 mi Beats the official forecast by 28% 14 mph Behind the official forecast by 159%
1 Google DeepMind 24 h 27 29 mi Beats the official forecast by 20% 14 mph Behind the official forecast by 20%
1 Google DeepMind 36 h 25 34 mi Beats the official forecast by 21% 16 mph Beats the official forecast by 5%
1 Google DeepMind 48 h 23 36 mi Beats the official forecast by 17% 19 mph Beats the official forecast by 4%
1 Google DeepMind 60 h 21 39 mi Beats the official forecast by 17% 21 mph Beats the official forecast by 16%
1 Google DeepMind 72 h 19 43 mi Beats the official forecast by 6% 22 mph Beats the official forecast by 18%
1 Google DeepMind 96 h 15 59 mi Beats the official forecast by 9% 27 mph Beats the official forecast by 7%
1 Google DeepMind 120 h 11 88 mi Beats the official forecast by 17% 35 mph Behind the official forecast by 1%
2 WeatherNext 3 12 h 29 17 mi Beats the official forecast by 33% 13 mph Behind the official forecast by 142%
2 WeatherNext 3 24 h 27 27 mi Beats the official forecast by 23% 13 mph Behind the official forecast by 9%
2 WeatherNext 3 36 h 25 33 mi Beats the official forecast by 22% 17 mph Beats the official forecast by 1%
2 WeatherNext 3 48 h 23 39 mi Beats the official forecast by 10% 19 mph Beats the official forecast by 3%
2 WeatherNext 3 60 h 21 41 mi Beats the official forecast by 12% 22 mph Beats the official forecast by 12%
2 WeatherNext 3 72 h 19 43 mi Beats the official forecast by 7% 26 mph Beats the official forecast by 3%
2 WeatherNext 3 96 h 15 53 mi Beats the official forecast by 18% 34 mph Behind the official forecast by 17%
2 WeatherNext 3 120 h 11 97 mi Beats the official forecast by 9% 40 mph Behind the official forecast by 15%
2 NHC official 12 h 29 26 mi — 6 mph —
2 NHC official 24 h 27 36 mi — 12 mph —
2 NHC official 36 h 25 42 mi — 17 mph —
2 NHC official 48 h 23 44 mi — 20 mph —
2 NHC official 60 h 21 46 mi — 25 mph —
2 NHC official 72 h 19 46 mi — 27 mph —
2 NHC official 96 h 15 64 mi — 29 mph —
2 NHC official 120 h 11 106 mi — 35 mph —
4 HMON 12 h 27 45 mi Behind the official forecast by 13% 13 mph Behind the official forecast by 299%
4 HMON 24 h 27 46 mi Behind the official forecast by 1% 20 mph Behind the official forecast by 91%
4 HMON 36 h 27 57 mi Behind the official forecast by 6% 26 mph Behind the official forecast by 82%
4 HMON 48 h 27 83 mi Behind the official forecast by 41% 32 mph Behind the official forecast by 82%
4 HMON 60 h 27 105 mi Behind the official forecast by 65% 33 mph Behind the official forecast by 59%
4 HMON 72 h 27 138 mi Behind the official forecast by 126% 33 mph Behind the official forecast by 59%
4 HMON 96 h 23 181 mi Behind the official forecast by 142% 32 mph Behind the official forecast by 63%
4 HMON 120 h 19 207 mi Behind the official forecast by 71% 33 mph Behind the official forecast by 32%
5 HAFS-B 12 h 39 29 mi Beats the official forecast by 13% 15 mph Behind the official forecast by 204%
5 HAFS-B 24 h 37 40 mi Behind the official forecast by 10% 19 mph Behind the official forecast by 99%
5 HAFS-B 36 h 35 51 mi Behind the official forecast by 13% 24 mph Behind the official forecast by 83%
5 HAFS-B 48 h 33 65 mi Behind the official forecast by 32% 27 mph Behind the official forecast by 78%
5 HAFS-B 60 h 31 87 mi Behind the official forecast by 56% 29 mph Behind the official forecast by 59%
5 HAFS-B 72 h 29 118 mi Behind the official forecast by 107% 31 mph Behind the official forecast by 67%
5 HAFS-B 96 h 25 165 mi Behind the official forecast by 101% 33 mph Behind the official forecast by 62%
5 HAFS-B 120 h 21 226 mi Behind the official forecast by 89% 38 mph Behind the official forecast by 42%
6 HAFS-A 12 h 39 31 mi Beats the official forecast by 4% 14 mph Behind the official forecast by 209%
6 HAFS-A 24 h 37 39 mi Behind the official forecast by 8% 18 mph Behind the official forecast by 84%
6 HAFS-A 36 h 35 60 mi Behind the official forecast by 33% 20 mph Behind the official forecast by 60%
6 HAFS-A 48 h 33 70 mi Behind the official forecast by 59% 21 mph Behind the official forecast by 35%
6 HAFS-A 60 h 31 91 mi Behind the official forecast by 91% 24 mph Behind the official forecast by 28%
6 HAFS-A 72 h 29 115 mi Behind the official forecast by 145% 26 mph Behind the official forecast by 29%
6 HAFS-A 96 h 25 159 mi Behind the official forecast by 155% 31 mph Behind the official forecast by 43%
6 HAFS-A 120 h 21 197 mi Behind the official forecast by 111% 35 mph Behind the official forecast by 36%
7 Canadian GEM 12 h 19 52 mi Behind the official forecast by 93% 27 mph Behind the official forecast by 664%
7 Canadian GEM 24 h 18 72 mi Behind the official forecast by 94% 28 mph Behind the official forecast by 226%
7 Canadian GEM 36 h 17 84 mi Behind the official forecast by 92% 32 mph Behind the official forecast by 170%
7 Canadian GEM 48 h 16 104 mi Behind the official forecast by 124% 34 mph Behind the official forecast by 130%
7 Canadian GEM 60 h 15 123 mi Behind the official forecast by 124% 34 mph Behind the official forecast by 89%
7 Canadian GEM 72 h 14 139 mi Behind the official forecast by 116% 37 mph Behind the official forecast by 77%
7 Canadian GEM 96 h 12 124 mi Beats the official forecast by 13% 45 mph Behind the official forecast by 108%
7 Canadian GEM 120 h 10 152 mi Ties the official forecast 54 mph Behind the official forecast by 90%
7 GEFS mean 12 h 39 34 mi Behind the official forecast by 14% 21 mph Behind the official forecast by 381%
7 NAVGEM 12 h 37 47 mi Behind the official forecast by 72% 32 mph Behind the official forecast by 609%
7 GEFS mean 24 h 37 49 mi Behind the official forecast by 26% 24 mph Behind the official forecast by 155%
7 NAVGEM 24 h 35 65 mi Behind the official forecast by 84% 32 mph Behind the official forecast by 225%
7 GEFS mean 36 h 35 65 mi Behind the official forecast by 38% 27 mph Behind the official forecast by 105%
7 NAVGEM 36 h 33 101 mi Behind the official forecast by 131% 30 mph Behind the official forecast by 113%
7 GEFS mean 48 h 33 87 mi Behind the official forecast by 71% 29 mph Behind the official forecast by 94%
7 NAVGEM 48 h 31 137 mi Behind the official forecast by 183% 25 mph Behind the official forecast by 51%
7 GEFS mean 60 h 31 110 mi Behind the official forecast by 84% 31 mph Behind the official forecast by 67%
7 NAVGEM 60 h 29 159 mi Behind the official forecast by 195% 17 mph Beats the official forecast by 17%
7 GEFS mean 72 h 29 132 mi Behind the official forecast by 112% 33 mph Behind the official forecast by 67%
7 NAVGEM 72 h 27 185 mi Behind the official forecast by 216% 15 mph Beats the official forecast by 37%
7 GEFS mean 96 h 25 170 mi Behind the official forecast by 99% 37 mph Behind the official forecast by 77%
7 NAVGEM 96 h 23 235 mi Behind the official forecast by 192% 20 mph Beats the official forecast by 21%
7 GEFS mean 120 h 21 194 mi Behind the official forecast by 65% 43 mph Behind the official forecast by 59%
7 NAVGEM 120 h 19 297 mi Behind the official forecast by 143% 26 mph Beats the official forecast by 3%
10 GFS 12 h 39 36 mi Behind the official forecast by 14% 15 mph Behind the official forecast by 237%
10 GFS 24 h 37 49 mi Behind the official forecast by 25% 20 mph Behind the official forecast by 109%
10 GFS 36 h 35 72 mi Behind the official forecast by 50% 23 mph Behind the official forecast by 74%
10 GFS 48 h 33 108 mi Behind the official forecast by 114% 25 mph Behind the official forecast by 64%
10 GFS 60 h 31 148 mi Behind the official forecast by 162% 27 mph Behind the official forecast by 42%
10 GFS 72 h 29 183 mi Behind the official forecast by 210% 28 mph Behind the official forecast by 43%
10 GFS 96 h 25 253 mi Behind the official forecast by 196% 31 mph Behind the official forecast by 52%
10 GFS 120 h 21 300 mi Behind the official forecast by 129% 36 mph Behind the official forecast by 38%
10 HWRF 12 h 27 38 mi Behind the official forecast by 24% 15 mph Behind the official forecast by 334%
10 HWRF 24 h 27 50 mi Behind the official forecast by 24% 21 mph Behind the official forecast by 98%
10 HWRF 36 h 27 71 mi Behind the official forecast by 41% 28 mph Behind the official forecast by 96%
10 HWRF 48 h 27 96 mi Behind the official forecast by 77% 34 mph Behind the official forecast by 92%
10 HWRF 60 h 27 111 mi Behind the official forecast by 88% 36 mph Behind the official forecast by 73%
10 HWRF 72 h 27 142 mi Behind the official forecast by 142% 32 mph Behind the official forecast by 63%
10 HWRF 96 h 23 167 mi Behind the official forecast by 128% 31 mph Behind the official forecast by 59%
10 HWRF 120 h 19 173 mi Behind the official forecast by 72% 31 mph Behind the official forecast by 22%
12 UKMET 12 h 18 41 mi Behind the official forecast by 9% 26 mph Behind the official forecast by 670%
12 UKMET 24 h 17 51 mi Behind the official forecast by 35% 30 mph Behind the official forecast by 237%
12 UKMET 36 h 16 55 mi Behind the official forecast by 32% 32 mph Behind the official forecast by 151%
12 UKMET 48 h 15 85 mi Behind the official forecast by 89% 35 mph Behind the official forecast by 119%
12 UKMET 60 h 14 109 mi Behind the official forecast by 121% 38 mph Behind the official forecast by 89%
12 UKMET 72 h 13 142 mi Behind the official forecast by 166% 38 mph Behind the official forecast by 93%
12 UKMET 96 h 11 198 mi Behind the official forecast by 99% 46 mph Behind the official forecast by 99%
12 UKMET 120 h 9 295 mi Behind the official forecast by 41% 54 mph Behind the official forecast by 53%

How we grade

ECMWF AIFS, ECMWF HRES, ECMWF ENS mean, AI GFS: No graded forecasts yet.

Track map

The solid line is the best track. Dashed lines are official forecast tracks from a few advisories, from the issue position out through the leads stored in that advisory. Thinner lines are model tracks from NHC’s public a-deck for those same advisories, and only when that file actually contains them. Google DeepMind and WeatherNext 3 are ensemble-mean tracks from Weather Lab, drawn on those same advisories. ECMWF AIFS, ECMWF HRES, and the ECMWF ENS mean are open-data cyclone tracks, drawn the same way. AI GFS is NOAA’s AIGFS tracker, drawn when that file is still posted.

8°N 14°N 20°N 26°N 139°W 129°W 118°W 107°W
Map of Tropical Storm Odalys. The best track, where the storm actually went, has 45 positions. 4 official forecast tracks are drawn. Model lines: GFS, GEFS mean, HAFS-A, HAFS-B, HWRF, HMON, NAVGEM, Google DeepMind, WeatherNext 3. Latitude and longitude, scaled by the cosine of the mid-latitude so a degree of longitude is not stretched.
  • Actual track (45 positions)
  • NHC official Sep 20, 06:00 UTC
  • NHC official Sep 22, 18:00 UTC
  • NHC official Sep 25, 06:00 UTC
  • NHC official Sep 27, 18:00 UTC
  • GFS
  • GEFS mean
  • HAFS-A
  • HAFS-B
  • HWRF
  • HMON
  • NAVGEM
  • Google DeepMind
  • WeatherNext 3

The chart measures how far the track was off in statute miles, the same miles as the home card. It includes 12, 24, 36, 48, 60, 72, 96, and 120 hours when a paired fix exists. The overall rank uses 24, 48, 72, and 120 hours. Positions are not interpolated.

How far the track was off

26 mi 12 h 36 mi 24 h 42 mi 36 h 44 mi 48 h 46 mi 60 h 46 mi 72 h 64 mi 96 h 106 mi 120 h
Average official track miss in miles by how far ahead. 12 hours 26 miles over 29 forecasts. 24 hours 36 miles over 27 forecasts. 36 hours 42 miles over 25 forecasts. 48 hours 44 miles over 23 forecasts. 60 hours 46 miles over 21 forecasts. 72 hours 46 miles over 19 forecasts. 96 hours 64 miles over 15 forecasts. 120 hours 106 miles over 11 forecasts.
LeadHow many hours ahead this forecast was made. This map shows 12, 24, 36, 48, 60, 72, 96, and 120 when a paired fix exists. Overall rank uses 24, 48, 72, and 120. The technical name is lead. CasesHow many past forecasts went into this average. A forecast counts only when the best track has a fix at that exact time. The technical name is cases. Track errorAverage how far the predicted storm path was from where it actually went, in statute miles. Lower is better. The technical name is track error. Intensity errorAverage how far off the predicted wind speed was, in miles per hour, on this page. Lower is better. The file stores knots. The technical name is intensity error. Forecast max windAverage top wind in the official forecast, in miles per hour, on the paired cases. The file stores knots. The technical name is forecast maximum wind. Best-track windAverage top wind where the storm actually went, in miles per hour, at the same times. The file stores knots. The technical name is best-track wind.
12 h 29 26 mi 6 mph 82 mph 81 mph
24 h 27 36 mi 12 mph 81 mph 83 mph
36 h 25 42 mi 17 mph 80 mph 86 mph
48 h 23 44 mi 20 mph 77 mph 89 mph
60 h 21 46 mi 25 mph 73 mph 92 mph
72 h 19 46 mi 27 mph 70 mph 94 mph
96 h 15 64 mi 29 mph 70 mph 99 mph
120 h 11 106 mi 35 mph 66 mph 100 mph

Miles are statute miles. Wind is miles per hour. The National Hurricane Center file stores wind in knots. One knot is about 1.15 miles per hour. Nautical miles are in the technical details.

How we grade

Top wind

12 h 24 h 36 h 48 h 60 h 72 h 96 h 120 h
Top wind, official forecast versus where the storm went. At 12 hours the forecast averaged 82 mph and where the storm actually went averaged 81 mph. At 24 hours the forecast averaged 81 mph and where the storm actually went averaged 83 mph. At 36 hours the forecast averaged 80 mph and where the storm actually went averaged 86 mph. At 48 hours the forecast averaged 77 mph and where the storm actually went averaged 89 mph. At 60 hours the forecast averaged 73 mph and where the storm actually went averaged 92 mph. At 72 hours the forecast averaged 70 mph and where the storm actually went averaged 94 mph. At 96 hours the forecast averaged 70 mph and where the storm actually went averaged 99 mph. At 120 hours the forecast averaged 66 mph and where the storm actually went averaged 100 mph.
  • Official forecast maximum wind
  • Best-track maximum wind

Source: National Hurricane Center / NOAA. The operational b-deck and the public a-deck are U.S. government works in the public domain. Google DeepMind and WeatherNext 3 ensemble means are experimental forecasts from Google DeepMind Weather Lab. Forecasts older than 48 hours are used under CC BY 4.0. ECMWF AIFS, ECMWF HRES, and ECMWF ENS tracks are ECMWF open data, used under CC BY 4.0. AI GFS tracks are a NOAA AIGFS product. Forecast Scoreboard is not affiliated with NHC, NOAA, Google, or ECMWF.