We replayed last winter's bulletins and weather on 39 backcountry slopes, every day from December to mid-April. Here's what a season looked like from the slope's side, and how many weekends had something good.
Most of us read the avalanche forecast one day at a time. We read it for a region, then translate it in our heads to the slope we're thinking about. We wanted to see a whole season of that translation at once.
So we took 39 saved zones (real lines people ski, all from one Snow Doctor account, ours) and replayed winter 2025-26 against them. For every day from December 1 to April 15, we took the bulletin in force that morning and the weather at each slope's own height, and asked simple questions. Was the danger Considerable or higher at that slope's elevation? Was a listed avalanche problem pointed at its band and aspect? Had the last two days loaded it? Did it warm up fast? That's 5,301 slope-days across eight areas:
It's not a study and it isn't peer-reviewed. It's one winter, on slopes one person happened to save. But a season seen all at once shows things a single day doesn't.
How often the danger was Considerable or higher at each slope's own elevation band, by month:
Early season and the March storms were busier than the middle of winter. If you plan your big objectives for "deep winter", last season's quietest stretch was January and February.
Half of all alpine slope-days were Considerable or higher. Below treeline, fewer than one in ten were. High or higher showed up on about one slope-day in ten overall, three quarters of them in the alpine.
This was the surprise. We expected the bulletin's avalanche problems to sort our slopes by which way they faced: the north-facing ones inside the problem, the south-facing ones outside it. Mostly, they didn't.
In the Sea-to-Sky bulletin last winter, 58% of listed problems covered all eight aspects. On 65% of days, every one of our Garibaldi slopes sat inside at least one listed problem, whatever direction it faced. The same was true in Spearhead.
When problems did leave some slopes out, it was usually by elevation: a problem listed for the alpine and treeline, with the below-treeline slopes outside it.
How often each kind of avalanche problem was listed, as a share of forecast days:
| Area | Storm slab | Wind slab | Persistent slab | Wet loose | Cornice |
|---|---|---|---|---|---|
| Sea-to-Sky | 46% | 38% | 12% | 32% | 22% |
| North Shore and Sky Pilot | 38% | 29% | 0% | 35% | 5% |
| Coquihalla | 38% | 37% | 8% | 30% | 11% |
| Duffey | 25% | 53% | 29% | 29% | 22% |
| Rogers Pass | 44% | 39% | 40% | 7% | 10% |
The coastal ranges were storm-and-wind country: new snow, rebuilt by wind, settling out within days. The North Shore didn't see a persistent slab listed all winter. Move inland and it changes. In the Duffey a persistent slab was listed on more than a quarter of days, and at Rogers Pass on 40%.
We wanted a simple weather-only flag for "this slope has just been loaded": a lot of new snow or rain in the last two days, or strong wind with fresh snow to move. To check it, we compared it with what forecasters listed. On the days the flag fired, how often was a storm slab or wind slab listed on that slope's band?
Adding wind read 80 metres above the ground changed almost nothing. Adding wind read at ridgetop height made the flag agree with forecasters 92% of the time. With ridgetop wind in it, the flag fired on about half of alpine slope-days and about one in ten at treeline. The alpine was being loaded most of the winter.
We also flagged rapid warming: a high above freezing at the slope's height, at least five degrees warmer than the coldest of the three days before.
It was rare all winter and common by April. And when it happened, forecasters had a wet loose or wet slab problem listed on that band 70% of the time, against 23% on any given day.
Bulletins give a rating for today and the next two days. We compared each "tomorrow" rating with the rating the next morning's bulletin actually gave for that day. They matched 71% of the time. About three times in ten, the rating changed overnight.
We used this winter to set how My Zones reads every saved slope: an avalanche list that starts at 0 and takes a point off for each thing working against the slope, and a weather score that adds and subtracts for snow, sun, cloud, wind, rain and warmth. Here's how every slope would have looked on the default values, the colours its pin would have shown on the map.
The weather was on the skier's side more often than not: light blue or better on 56% of slope-days, and grey on only 12%. The trees did best, light blue or better on 69% of treeline days, out of the wind and with contrast when the cloud came in.
The weather score is personal. In the app you pick a starting set (storm days, clear days or whatever skis best) and change any value after that. The same winter, three ways:
Most of us tour on weekends, so we pulled out every Saturday and Sunday from December 1 to April 15: 19 weekends, 38 days, on the 36 slopes on the South Coast and South Coast Inland (everything but Rogers Pass). We called a slope's day quiet on the avalanche list when its pin was 0 or -1, and good on weather when it was light blue or better.
How many of the 36 slopes were quiet on the avalanche list and light blue or better on weather, each Saturday and Sunday:
Two things stand out. Weekends weren't unlucky: they looked like any other day of the week. And there was almost always somewhere. Even on weekends when the alpine was loaded and grey, a treeline or below-treeline slope was usually quiet on both counts: treeline slopes managed it on 41% of weekend days, alpine slopes on 12%. With blue weather instead of light blue, the same-day count drops to 20 of 38 days, still on 14 of 19 weekends.
Quiet on the avalanche list doesn't mean safe: it means none of the forecast items we check applied, and the slope questions were blank on these slopes. Read the bulletin, look at the snow, and make the call with your crew.
One winter is one winter. A different season, or a different set of slopes, would give different numbers. What we'd expect to hold is the shape: height first, aspect less than you'd think on the coast, wind measured high, and spring warming as a real signal.
We used this run to rebuild how My Zones reads your terrain. Each zone now gets an avalanche list (what counts against it, starting at 0 and only going down) and a separate weather score that you set yourself. The loading and warming flags above show under that list as a yes or no, at the thresholds that agreed with forecasters most often. They take no points off: forecasters had usually listed the matching problem already. Every rule is in Every Spot You Ride, Read Two Ways Against Today's Forecast.
None of this replaces reading the bulletin, or what you see on the snow. It's what a season looked like when you line the forecast up against the slopes you actually ski.