Forecast evidence / Before you book
How much trust should you put in a winter outlook?
Use seasonal forecasts to compare possibilities. Before committing money, check how the source performed for that location, variable, season and lead time.
Our current accuracy status
We have not established a resort-level accuracy score for the seasonal forecasts displayed here. The site publishes sourced regional temperature and precipitation outlooks. It does not have verified evidence that these predict snowfall totals, opening dates or the quality of a particular ski week.
Source checks, update timestamps and expiry controls verify which forecast we display. They do not demonstrate that the weather prediction will be correct.
The older snowfall model’s accuracy claims remain withdrawn while its underlying records are reviewed. Read the historical-data status →
U.S. / NOAA CPC
Check the official track record
NOAA’s verification tool offers maps and charts. Choose temperature or precipitation, the seasonal period and a score; inspect the result for your region.
A skill score compares performance with a reference. NOAA’s Heidke score measures improvement over chance; zero means no improvement. It is not the percentage of forecasts that were right, and it should not be compared numerically with a percent-correct score.
Canada & global / CanSIPS
Read skill alongside probability
In the official viewer, match your location, variable, forecast period and lead time, then inspect the skill map. Lead time is the gap between forecast creation and the start of the predicted season.
The Canadian guide describes percent-correct maps: how often the predicted category matched historical outcomes. Its reliability diagrams compare forecast probabilities with observed frequencies. A large forecast probability alone does not establish high skill. Check the model version and reference years too; an older system’s score cannot automatically validate a newer forecast.
Open Canadian forecast and skill maps ↗
Find global forecast products ↗ · Skill and reliability guide ↗
Turn that evidence into a trip decision
- Match the forecast to your trip. December–February is a different target from February–April. A forecast issued one month ahead is a different test from one issued nine months ahead.
- Read temperature and precipitation separately. A wetter regional season does not establish more mountain snow. A seasonal average cannot pick your best ski week.
- Keep uncertain choices flexible. Compare destination costs and cancellation deadlines. Do not treat a weak or unverified signal as proof that a more expensive trip will deliver better skiing.
- Recheck as departure approaches. Use short-range weather, operator reports and webcams to assess actual conditions and operations.
What would justify an accuracy claim here?
A documented test would need archived forecasts issued before the outcome, independent observed results, matching locations and calendar windows, a stated comparison baseline, and enough cases to show uncertainty. Temperature, precipitation and mountain snowfall would need separate evaluation. We have not completed that validation for these mountain outlooks.
See forecast sources and coverage → · Return to the winter planner →