UKC

I haven’t had much need for serious avalanche decision making over the last couple of years. I’m sure the reasons for that will be apparent to anyone who’s been trying to winter climb in Scotland over that time! However, yesterday bucked that trend.

I wanted to open a discussion about such decision making and I offer our experience yesterday as a starting point. Maybe I'm not the only one who feels in need of a reminder?

Yesterday we climbed Lord Berkeley’s Seat on An Teallach. This involves a long approach up Lord’s Gully. We knew from the recent weather and the SAIS report (Torridon is close enough to be a reasonable proxy I think) that the gully would present potential avalanche risk. We’d identified this prior to heading out and planned to assess conditions in the gully and make a decision then. Two of our team of three are regular partners, we are not averse to walking away from things if need be and have done so in the past. Our plan B was to ab in from the ridge but in retrospect this would have been a time consuming ball ache.

On arrival we found the gully to be pretty well filled with new snow, deposited over the last couple of days which was naturally of concern. We had a bit of a discussion and decided to proceed on the basis that:

• There was no evidence of a base of older snow on which new snow might slide.
• There was no evidence of a weak later in the snow pack. It was very homogeneous, I suppose because it had all essentially been deposited in the same event.
• There was no evidence of windslab, it just looked like fresh rather than wind re-distributed snow.

One thing we didn’t do, that I have done in the past and found to be helpful, is have a more formal discussion about risk and evidence, it was really just chatting.

We climbed the gully, sticking to the edges/rockier areas and spreading out so only one of us was on the more open, deeper sections at any one time. We weren’t avalanched but as everyone knows that doesn’t mean we got it right!

I’m curious to know other peoples thoughts on our decisions and to hear what approaches to decision making you use.

Link to the SAIS forecast for Sunday: https://www.sais.gov.uk/torridon/?report_id=12309

Oh and what a route! Good quality climbing in a spectacular setting with the added spice of a non-straightforward approach and descent. Lord Berkeley's Seat (VI 6)

Did you dig a test pit for snowpack stability?

We didn't dig a pit as such but did have a bit of a dig around at various points. I'm not entirely convinced of the value of digging pits as an 'on the hoof' tool for decision making. Doing them properly requires quite a lot of skill and they only give localised information. How many of us could say hand on heart that they could use the info they gleaned from a pit to make a solid call on whether to proceed or not?

Localised information is exactly what you want, not generalised.  It’s part of the decision making process, and I’d say not that complex or time consuming to do. 

Have you got any tips on performing a pit test without a shovel? I tend to dig a hole and have a look at layers etc but this isn't really the same as a proper pit test. Obviously the answer is 'carry a shovel' but I think most winter climbers are not particularly keen to add that weight to their bags.

A pit is only going to tell you about the snow pack at that exact location, not necessarily what it's like further up your route. Not saying there is no value in it, but it needs a bit of thought as to whether your pit is representative of the whole of your route. And a pit should really just be to confirm what you would already be thinking from the avalanche and weather forecasts and observations enroute. 

To DaveHK, the next days avalanche forecast has the observed snow conditions for the day you were climbing, which would be more useful as it's based on actual observations rather than a forecast. For Torridon it gave a moderate hazard, with a mention of windslab in sheltered locations NW through N to SE (what aspect was your route?). That said, the snow profile for yesterday (taken on Beinn Eighe) shows a softer layer (2 finger hardness) overlying a harder layer, a bit of a temperature gradient (which can maintain or grow instabilities), but both layers of rounded grains, which are a sign of a stabilizing pack.

Obviously you would have a better idea of the conditions than me, being stuck at home, but from what you describe it sounds like a reasonable call to me.

When making go / no-go decisions in avalanche terrain, each decision is a bet.  That's because you're making a decision that'll have real-life consequences without all the information you need to be sure how things will play out.  By nature it's gambling.

So you need to think about three things.  1.  Probability of triggering an avalanche.  2.  The consequences if you're caught.  3.  Your party's level of danger tolerance.

If probability x consequence is greater than your tolerance then it's a no-go.

Assessing probability of triggering a slide is complex and we all get it wrong sometimes because it involves dealing with all the vagaries of snowpack, layering, loading, spatial variation etc. so gather all the data points you can and plug those into your mental model of avalanche initiation for the relevant terrain.  Then allow some margin for human factors (commitment heuristic might be a relevant one in your example.)

Consequences are often more straightforward to assess.  You're looking for slope size, how much snow is on it, likely run-out, terrain traps etc.  Think about possible avalanche size and where you're likely to end up.

Danger tolerance?  Well that's up to you.  It's going to be different in a professional environment than for most amateur parties but that's the freedom we have.

Yes, I had a look at that earlier. The approach up the gully was NE facing.

For testing, you first need to think abut your test site.  Three rules:  1.  Safe.  2 Representative of the slope you're considering climbing on (wrt aspect, angle, altitude)  3.  Undisturbed.

Then think about what you're testing for.  What kind of avalanche are you expecting, given the conditions?  If it's a wet-snow avalanche, a pole probe test is a good one (and easy to do lots of as you move.)  For windslab, you're looking for an "upside-down" snowpack  i.e. a denser layer over a weaker one.  Testing resistance (there's a standard format for that which is probably better seen in a video than me trying to describe it here,) would be an appropriate test in that case.

A shovel compression test is a good way to identify weak layers in the pack.  Again, video will be better than description.

... and if you haven't got a shovel with you, a hand shear is not as repeatable but can also show up weak layers.

Reflecting on yesterday's conditions I think they were fairly unusual, or at least not the sort of stuff that is generally discussed in that it was essentially one big fall of snow on no base. My understanding is that in these circumstances there would be an initial period of instability then the snowpack would gradually stabilise, assuming there are no further changes? 

Does anyone have any views on the relevance of the lack of base? The state of the underlying ground was probably quite variable, higher areas might have been frozen but lower parts were definitely wet. 

Back in 96 we made the second ascent, having   a similar conundrum  with  deep snow and pockets of  wind slab in the approach gully,  we opted for moving together placing the occasional runner. Although my memory is a little hazy about the climb, i had always wanted to climb the face of Lord Berkely's seat since a youth gazing down  into the abyss  from this airy perch.

Out of interest did you get to the base of the route, clip into a belay and think 'We were lucky there...'

Obviously not anywhere near a fool proof assessment technique, but worth noting if you have been out in winter a lot, as you have. 

There was definitely an element of relief there but I didn't feel we'd taken a risk and got lucky and nothing else across the day changed my mind on that. I felt we'd engaged in a rational process and made a rational decision but the issue is just how rational can it actually be? A lot of experienced people get avalanched after feeling like they've made a rational decision.

I was discussing this topic with my GF recently and she made an interesting point about going up slopes / gullies to approach climbs. My logic is much like yours, and I would say intuitively most people would skirt the sides & avoid the deeper central part of a gully. Energy expenditure probably motivating this as much as the perceived safety aspect. The point made by my GF was that the edges are the locations most likely to trigger an avalanche (often can be seen in videos of skiers in couloirs / gully’s). I haven’t been able to back this up (haven't really looked) & it would be good to have expert / mountain professional opinions on this. 

Given how baseless the snowpack often is in Scotland, ploughing right up the middle of a gully would be hard work (or not really doable) in deep snow. Perhaps if this is what ur faced with on the day then it’s better not to push on when combined with the avi forecast? 

Good thinking about a retrospective look. 
Walking up, I was purposefully looking for clues from the loch.

We waited to have daylight and view of the gully before going underneath it.

From our gear up point, with the snow around, I gathered that there was going to be a fair amount of snow but we had a couple of words as to how homogeneous it seemed.

while trail breaking, during pauses and gulping for air, we decided on a reasonably safe route and decided, likely wisely, to avoid the narrows via a ramp.

You are right we should perhaps have really stopped and formally chatted about it all: have we climbed too long together?

Or at least probe? Can be enough sometimes to see if there's a consolidated slip-surface hiding in the snowpack.

I've been waiting for the right moment to raise this...  

IMO if you haven't got a shovel and you are uncertain enough to be doing some rudimentary form of pit test, it's not worth going. If you were to be avalanched, how would you dig your pal(s) out without a shovel?

Lets not even mention the allergy climbers have to transceivers...

If you get some energy into the snowpack to begin internal melt and refreeze, then yes - the snowpack will stick to the ground. Like all fresh snowfalls what you want are some sunny, warm days to produce some meltwater which can percolate and refreeze at the base overnight. That meltwater generally remains frozen, even during 'warm' days, because snow is a great thermal insulatior, and the (shortwave) radiative energy flux (sunlight) is very low in the winter in the Northern Hemisphere (which FYI, can penetrate c. 10m into optically clear snow).

Avalanches in this set of conditions generally have a mode of full-depth failure - i.e. between the snow and the ground. But it also depends on the consolidation levels of the snow - a pile of fluff doesn't have the weight behind it to slide as a cohesive block - but a snowpack that itself is consolidated, and isn't frozen onto the ground (especially when the ground is already frozen) has reasonably high avalanche potential. Think of a brick on a slide, vs a pile of sand on slide. 

If you hadn’t got a plum prize in your sights and instead were just out for a gander would you have chosen to go into a sheltered aspect c.24hrs after the end of a big storm with no to minimal temperature fluctuations since, nor significant wind reversal, to wander up a grade I (leaving aside the physicality of trench war fare)?

Not ment as criticism, but as a counterpoint for thought/discussion…..

For those that have answered if the party concerned hadn’t been seasoned Highland based regulars and posters but newbies going up the atmospheric grade I would you have provided the same response?  Or would it have been very quickly a pile on.

Yours having triggered a massive 1.5m full depth 200m crown wall (I managed to step off the top of it) in fresh conditions when ‘young and stupid’. Then 20years later triggered a small one 20ft across with a 15cm crown wall when ‘old and wise’ with the thought process of ‘that won’t avalanche as its shallow and the top of the dwarf pines are stick out’,  and then watched pile into a small scoop 4x3x0.5m that wouldn’t even preserve your modesty for a number 2.

Don’t want a number 3!

As Huck Finn says "the statements was interesting but tough"

There is definitely a strong element of risk vs reward in this which can affect decision making. I think I can say quite honestly that if I was out for a potter around I would not have gone up that gully. How much of that would be motivated by risk and how much just plain unpleasantness I'm not sure. It's maybe a useful question to ask yourself on the day though?

The presence of others also has an effect.

I'm on 3 and really, really don't want there to be a 4!

Yes this is always how I've understood snowpits - they can tell you if a slope isn't safe to go on, but just because you get good results from the pit doesn't necessarily mean the slope is safe?

I've never really considered digging one in Scotland, and I am curious to know if any folk on here do it as a part of their decision-making (particularly climbers?). I'm lucky enough to live up North and have a passion for sacking off uni, which makes it easier to adopt a very conservative approach given I don't know loads, and if I have any concerns about a slope, I'll just avoid it. Definitely easier to do this mentally when I'm only two hours away and can just come back next week! 

Indeed, I forgot to say I probably would have gone for ‘it’s fresh on no base , so let’s sneak up the side’. Even though on a gander, or with people keen for a low grade romp would have said nah…you can I’m not.

There is also the dichotomy of trying to remove avalanche risk to zero, yet accept that on route rounded, flared, blind, buried and iced /rimed cracks might mean that the fall potential was longer/more fraught than you might be comfortable with.

Having listened in to our guide talk to one of the founders of the avalanche heuristics paradigms on a walkout from Ben Nevis, it was interesting to hear the take on snow pits. 

I can't remember the person's name but they said they had experiments where they dug a grid of about 100 snow pits across an area on a slope and the results were so varied as to make the results of any single pit pointless. And that was on a Canadian slope where the topography wasn't as varied as our rocky gullies where windslab can be very local and conditions of thaw/freeze change rapidly with altitude. This is anecdotal as I can't find the actual references, but I was also told the experiment had been repeated in Scotland with similar results.

 

There is a photo floating around of this test. It's on an essentially uniform slope and the results were totally all over the place.

A hasty pit can be used as a quick tool to gauge broad (or very narrow) conditions.on the day, this would be alongside keeping track of what the wind, precipitation and temperature are doing. Most planning should take place before going out, looking at weather and snowfall history, terrain etc... 

Finally the churning in your stomach is usually a good sign you're in the wrong place, even if it is ok, it's much better to be conservative and cautious and make it out climbing another day.

​​​

For me if it feels dodgy then it is dodgy, sounds like it felt dodgy!

It's great to be well informed on snow science/forecasts etc. but when you do have significant experience I feel the gut instinct is the most powerful indicator - obviously not much help if your gut is dodgy!

Does anyone have any views on the relevance of the lack of base? The state of the underlying ground was probably quite variable, higher areas might have been frozen but lower parts were definitely wet. 

If there's no old snow under the new then for it to avalanche it either has to slide on the ground or for there to be a weak layer within the new snow.

I'd be thinking about the ground in relation to anchoring (one of the five As.)  What's the bed surface?  A boulder filled gully?  Steep grass?  A rock slab with water running down it?

To identify weak layer in the new snow we're back to a shovel compression test and looking at layer resistances.

Sorry for the blunt reply but NO!

If you can't explain your decision in terms of terrain (Angle, Aspect, Altitude, Appearance, Anchoring) in relation to the avalanche hazard on the day then you've got nothing upon which to base a decision.  Really - nothing.

And experience is, unfortunately, a poor teacher when it comes to avalanches.  Most of the time, most of the snow is quite safe so experience usually teaches us that what we're doing is ok.

I'm glad you had a good day out, and clearly there wasn't an avalanche. However, I think there was a failure in the planning process.

The SAIS report gives considerable on the NE and E aspects above 700m. This is spot on for the approach gully you intended to use. There were plenty of lower hazard areas of Scotland you could have visited on Saturday. I took my guests to climb in Glencoe on Saturday because it was forecasted for moderate risk. I wouldn't have chosen to travel all the way to An Teallach to climb in a higher hazard area.

Digging snow pits and assessing snow conditions on the day isn't a good way to make decisions. You get a tiny snapshot of the snow which doesn't tell you anything about the whole slope. It's even less useful at the bottom of a gully to provide data about the top. My one caveat would be if you turned up on given forecast and found far more reactive snow than anticipated. This would obviously affect my decisions. The forecast does occasionally get things wrong.  

Looking from the outside, and without wanting to sound critical, I think you had already formed a plan to climb a specific route and once you got to the corrie, a case of plan devotion bias and unwillingness to change your plan probably came into play. Plenty of other heuristics you might have fallen foul of too such as groupthink and resource scarcity. 

Use the BAA framework 'key point' to identify where honest group conversations should take place and make sure you have them.

Personally for work and personal climbing, I just go wherever the forecast is lowest and climb on safer aspects. Trying to outsmart the hazards isn't worth the risk even if the routes are really good. Theres always something reasonably safe you can do. 

This is definitely something to be aware of but I don't think we did that on Sunday. We had a plan B (abbing in) and were quite willing to follow it but decided it wasn't necessary having assessed the gully.

I think this was probably true, and again, I think this comes down to a planning error. It's not the best plan B and I imagine there was some conscious or unconscious decision making being made because no party members really wanted to slog up the ridge to the top.

I think it's preferable to pick a lower hazard area, have a selection of routes in mind, approach with an open mind, and choose the best option when you get to the corrie. Convoluted plan Bs never seem appealing once you get to the bottom of a route. 

I haven't read the whole thread but it sounds to me like you made good decisions on the ground.

If you have a read of the following day's Torridon forecast you will see that the observed hazard was Moderate https://www.sais.gov.uk/torridon/?report_id=12314

Looking at the Torridon blog for Sunday you'll see that the forecaster found better stability than expected which ties in perfectly with your snow observations https://torridonblog.sais.gov.uk/2025/01/a-pleasant-winter-day/

I think this is starting to touch on confirmation bias. Plus, getting away with it is the thin end of the wedge for the normalisation of deviance and trying to get away with it in again.
 

The problems or potential problems I've had with avalanches are mostly my own problems. I've been caught out once or twice where a huge dump of snow has happened during a day and we had to descend in dangerous conditions but I'm not counting that as my problem.

I think it boils down to- no matter what- I was going in to do a route in that venue.

'I've walked in X km and I'm prepared to risk it as once we are on the route it will be fine.'

First instance- early days, first few routes in early 90's. Aonach Mor East Face. Bollocks/ waist deep snow, very sunny trudging in thermal top across to the routes. Even thought it was early days I can remember feeling quite uneasy. 

Second- me and mate soloing Central Gully (Ben Lui) in quite hot, late season weather- maybe April? Luckily when the cornice collapsed we were above the narrow bit and had the opportunity to scuttle as fast as we could to the side as the big blocks tumbled down the upper bowl towards us.

I think it's a form of cognitive dissonance- you can convince yourself you've done a bit of technical climbing so, walking is fine, and hopefully it won't bite you on the arse. 

PS I've been in quite a few other circumstances that could have resulted in an avalanche but these jumped out at me as relevant.

No problem but I disagree - you have something very powerful to go on. Something that very often (in my experience) cuts through the heuristic traps that are so easy to fall into even when you have done plenty of assessment in real terms, that has led you to decide to press on.

My point was meant to be that when it feels wrong it usually is and you should listen to that, which clearly is something to base a decision on - same as in many other situations in life.  Obviously it is far more valuable if you know why it feels wrong and was not trying to imply snow science/forecasts are superflous!

Also I strongly disagree on second point, experience is the best teacher.

I agree with this. When I saw the SAIS report for Sunday my first thought was that maybe we'd made a good assessment of the snow pack but I quickly realised that was all too easy a conclusion!

In reply to 

Just to play the devil's advocate for a moment and to be absolutely clear I am not suggesting this is the correct approach. It is however a mindset that many climbers adopt.

The avalanche risk on Sunday was extremely variable. The amount of snowfall in different areas varied quite significantly. The wind had done it's usual crazy stuff and put snow in all sorts of places, not all of them predictable. A forecast is just that, it's not a prediction and the difference between forecast and observed conditions in Torridon bears this out.

The advice most of us would give is that if conditions are worse than the forecast (whether it be snow conditions or weather) one should adapt or scale back plans. Why should that not also work the opposite way if conditions are better than forecast? After all, surely those on the ground are best placed to make a judgement?

What's wrong with this?

My first thought is that it places an awful lot of weight on ones judgement on the day and there are numerous things which can go wrong with that. I also think that doing this and getting away with it can lead to dangerous hubris.

Any other thoughts?

It's worth people remembering that An Teallach is not within the Torridon avalanche forecast area. It's geography and proximity to the sea can give quite different conditions to The Fannaichs or Torridon. 

To quote those more eloquent and informed than myself:

"Experience doesn’t necessarily correlate to expertise. The quality of the experience is paramount when gaining expertise."

We need to be honest with ourselves as to what our "experience" amounts too.

"If we look at how ski guides gain expertise with the snowpack in contrast to ski patrollers, we can observe interesting comparisons in the structure of their learning.

Ski patrollers and forecasters are avalanche hunters. Guides are avalanche avoiders. By hunting their quarry, patrollers learn to interpret and identify avalanche problems differently. When a patroller rises at dawn for a hundred days every year with the sole aim of triggering avalanches, they engage in a process that over time assists the development of expert intuition.

Conversely, ski guides routinely make consequential decisions in an ocean of uncertainty. Awareness of an avalanche problems’ existence and its distribution is sufficient information to plan their day, effectively avoiding the avalanche problem. A seemingly wise strategy. However, the lack of direct feedback that results from this strategy can inhibit the development of intuition."

It is easy to think that because we have not been avalanched, we are making good decisions, when in reality, we may just have been lucky.

I would also like to say thank you to DaveHK for both starting the thread and responding to comments in good grace and for LucaC for posting a "robust" challenge to Dave's decision making but in a constructive manner.

UKC at its best imho.

For those interested in this kind of think, this is where the above quotes come from and worth a read: https://avalanchegeeks.com/2023/11/07/guide-mortality-in-fat-tailed-domains...

That may be true in a kind learning environment.  If you can practice a skill repeatedly in a situation where it is safe to make mistakes, where there is clear, timely feedback on your actions and where it is made obvious to you how you can improve then experiential learning can be great.

Avalanche terrain is not that environment.  It is a terrible learning situation for experiential learning.  When we make a bad call, most of the time we get away with it and every time that happens it gradually builds our internal idea that what we're doing is ok.  Clear feedback is simply absent most of the time.  Nothing is telling us how we could have behaved differently to reduce our exposure to a hazard.  When we do get actual feedback things have already gone wrong.

The process that LucaM describes is the way to reduce risk.  Start with the avalanche forecast, modify it as you go with your own observations from the field, mentally map it onto the terrain in your plan then make your decision based on whether the reward is worth the risk.

I’d listen to Luca if I were you, his is most helpful reply on the whole thread. 

In my view it is best to make big picture decisions whilst planning your day, and not try to bend the rules to suit your desired objective.

My general rule is if the aspect I want to climb on is considerable I’ll go and find a different route/crag with a moderate forecast. 

I take it you mean this gully? If so the base is rocky scree ( big lumps).

Yes, that's the chap, Lord's Gully.

A great route in itself but a tad undergraded at the top since it stopped being choked with ice on a regular basis.

Having spent a depressing time in there in summer I would guess that with no base it wouldn't be your worst nightmare, unlike Constabulary Couloir which has too much grass and slab.

The right fork is given II but it looked more like IV on Sunday. Not many logs on here.

On the whole we were happy with our decisions but I thought it was a good opportunity to revisit a few things!

Definitely not a II when I did it a few years ago, very relieved to top out.

Quite right. 

This reminds me of being avalanched in Great Gully Craig yr Ysfa under similar new snow conditions on the final approach to the climb- on the side I managed to jump clear but my pal was carried several hundred metres- we survived but felt like a close call- there was no mobile signal.. at the break was a clear fracture about 1M high..

https://www.ukclimbing.com/logbook/crags/craig_yr_ysfa-530/great_gully_wint...

Thanks, I thought quite hard before posting.It's  difficult but important to keep reconsidering your decision making. And exposing your thoughts to public criticism obviously comes with a degree of trepidation!

I thought a lot of that was unnecessarily obscure but some of it tied in with a podcast I listened to recently that discussed (in part) the difference between complicated and complex problems.

Complicated problems are those with lots of elements but those elements are largely predictable and you can basically do the maths and formulate a sound respond.

Complex problems are where not all the variables are predictable or where there are hidden variables so just doing the maths won't necessarily lead to a good response.

Avalanche decision making is a complex problem but by thinking we can make decisions that will allow us to safely navigate avalanche terrain we're treating it as a complicated problem. If we keep doing this or try to make marginal calls we're exposing ourselves to serious consequences.

Thanks for posting this Dave. As galpinos said, this is UKC at its best.

I’ve been pondering similar questions for the last few months. After 20-odd years of Scottish winter climbing without any avalanche incidents, I’ve been caught out twice in the last couple of years. Both were very localised hazards; in hindsight, one where I’m deeply frustrated at myself for missing or ignoring warning signs, and one where I think luck played a genuinely larger role.

If Lord’s Gully had avalanched when you were in it, I think you’d look back on it with frustration at a bad decision rather than considering yourself unlucky.

My incidents have made me reflect on my own experience. As Adrian explained better than I can, the fact that I’ve done quite a lot doesn’t necessarily mean I’ve made good decisions. There are definitely times when I think I got away with bad decisions. Right now, I think I'm over-estimating the hazard and have a level of paranoia every time I'm on a snowslope. 

FWIW, for Sunday, I expected a high avalanche hazard on anything E/NE facing because of the storm-force WSW winds on Friday. I take Roger’s point about An Teallach not being Torridon, but the Torridon forecast confirmed my view of likely hazards and it didn’t seem like the weather had been that different between Torridon and Dundonnel. My starting point for planning Sunday was thinking of west-facing crags with west- or south-facing approaches. Aonach Beag and Druim Shionnach were also on the shortlist, but I’m glad we took the punt on a more adventurous day.

I don't mean ignore the Torridon forecast for An Teallach just be aware that it's not within the forecast area and conditions may be worse or better than suggested. 

It's quite likely that the time afforded us for such reflection would have been rather brief!

It won't directly answer the OPs question, but anyone posting on or viewing this thread might find this joint Met Office/SAIS talk on the 4th Feb interesting and relevant. https://www.ukclimbing.com/forums/premier_posts/met_office_sais_talk_4th_fe...

(not wanting to distract from the good discussions going on here, I thought it worth highlighting as it would provide an opportunity to ask questions to one of the people who actually write the avalanche forecast)

It's an interesting thought experiment, but I'm not sure how useful it often is in practice. Many climbers would accept huge (very serious) consequences if the probability was low. At the same time, no-one wants to die so if there's a high probability of a serious outcome, acceptance of risk is going to be low amongst nearly everyone. 

The crux of the matter is estimating probability, and it's because that factor is so difficult to assess that accidents happen. 

I'm glad this has been a good conversation about avalanche and risk. It takes a lot to reflect honestly about risk and more so to do it in a public forum. 

I think as winter climbers, we have really good opportunities to choose our terrain, aspect and altitude on a given day to reflect the forecasted avalanche conditions. I always err on the side of caution and in the planning phase for any day out, try and find somewhere to go with the lowest risk that will still give us a good day for whatever the experience and aspirations of the group. 

The west coast of Scotland has some great lower level crags and west facing cliffs which can be safer after storms from the south-west. I would always pick one of these rather than chancing it on a considerable aspect. 

If we plan effectively we can take a lot of the guesswork out of what the snow is doing by simply avoiding the problem.

If this thread has made anyone reevaluate their own avalanche avoidance practices, the SAIS also have some resources available: https://beaware.sais.gov.uk/resources/ and some learning material about the BBA process: https://beaware.sais.gov.uk/ both of which can help make sound decisions. 

The hard thing with avalanche assessment is that you do not get much feedback. When you learn to rock climb, the rope gives you feedback: if it gets stuck or pulls out your wires, you know that you need to work on your clipping. When it comes to planning a route in winter we don't have that luxury, also because, as mentioned by other users, it also depends on the group.

I've only been climbing for 6 years but have already been avalanched once (on a Low risk!) and have triggered a huge crack on the slab in Coire-an-Lochain. That day (Moderate), we were watching a pair heading straight up the slab towards Ewen's and were placing bets on when would they trigger a crack and yet even with our own precautions of staying close to the buttresses where the snow was more shallow, we ended up being the ones that probably got closest to danger.

It's hard to say how 'scientific' assessing these risks are as it depends so much on the group and many other factors. I think the best way to improve on assessing avalanche risk, since there's no real feedback from the snow (unless you trigger a big one), is for climbers to keep sharing opinions and generally just talk about it so we do not lower our guard.

Local knowledge also helps a lot. So, for example, in your case it might have been helpful to talk to guides who climb a lot in Torridon. There are some corries where even on a Low risk I'd be keeping my eyes open because I know from local guides that they are hotspots.

Let's keep up this thread - as folk have pointed out, it's UKC at its best.

A couple of things to add to some already good points: there’s a great deal of spatial variability in any given snowpack, especially Scotland, so use any form of shear test as additional evidence only, don’t use it as a green light to proceed. On the other hand take it as a warning if you’re seeing shooting cracks or blocks of slab ejecting from under your feet. 
 

Wind speeds in the same weather event can vary markedly and this reflects the density of wind slab. You might not see a visible ‘layer’ interface as such but there may enough differentiation to create an instability. 
 

To disrupt inherent biases, that we all have, make an agreement in advance to stop at certain points and honestly evaluate the situation. Pre-plan some sufficiently appealing alternatives - part and parcel of ‘key places’ that others have already mentioned. 
 

One of the biggest things that get climbers avalanched is not a lack of knowledge - it’s human behaviours. 

Full depth 'Gleitschnee (sliding snow)' avalanches are common in the Bavarian alps, normally on steep sunny grass slopes - I think light penetrates all the way down to the grass which then warms enough to cause melting at the base and whoosh, the whole pack avalanches or slumps at the least.  From the Bavarian avalanche warning service:-

"The entire snowpack slides on smooth ground such as grassy slopes or smooth rocky areas. High activity of sliding snow avalanches is typically associated with a thick snowpack with few or no weak layers. Avalanches can occur in both dry, cold and wet, 0°C isothermal snowpacks. It is almost impossible to predict the time of departure of sliding snow avalanches, although they are usually signalled by sliding snow cracks (so-called fish mouths).
Types of avalanches to be expected

Sliding snow avalanches: dry/cold and wet/0 °C-isothermal
 Almost exclusively spontaneous triggering. Artificial triggering is unlikely.

Where?
Mainly on smooth ground. In all exposures, but more often on south-facing slopes.
Position of the weak layer in the snowpack

At the transition from the snowpack to the ground.

Triggering mechanism

Sliding snow avalanches are triggered by the loss of friction on a moist and wet layer between the snowpack and the ground.

Duration

Days to months, triggering possible throughout the winter. Triggering can occur at any time of day or night. In spring, they usually occur later in the day.
How do I deal with them?
Recognising problems in the terrain

Although sliding snow cracks (fish mouths) are easy to recognise, it is virtually impossible to predict when they will be triggered. Triggering is also possible without the formation of sliding snow cracks.
Recommended behaviour

Do not stay in the vicinity of sliding snow cracks.

Yeah - glide avalanches are interesting because before they fail the support for the snow is from below rather than tension at the start-zone.  They're hard to control  because the obvious thing seems to be to drop a charge in the crack (fish mouth as mentioned above) but it does nothing.  There's some study been done about kicking out the toe with big charges (like >200kg!) with limited success.  The two patterns that seem to get reported is that they tend to release naturally just after the sun goes off them and they're on steep grassy slopes.

I think the good thing is that all the ones I've seen the remains of have been small (<50m across) and the slipped snow has remained in one piece, perhaps a bit folded up at the bottom.  So I doubt there's much burial risk. Having said that, the snowpack on those hasn't been terribly deep, only a foot or so.

Yes - this line of thinking is, in my view, the way forward and there is good value down this road.

When we seek certainty about what's going to happen we're fundamentally getting it wrong.  We'll almost never have certainty about the outcome of crossing a slope because we don't have all the relevant facts when we make the call.  (And that's before we even think about the many ways in which our own brains will make a mess of things.)  Whether we like it or not, that's simply the nature of the environment we've chosen to enter so we need to deal with it on nature's terms - not the ones we might wish for.  That's why I say that the only way I've found to come to grips with the decision making in avalanche terrain is to treat it as a series of bets.  Unfortunately gambling has a bunch of negative connotations but, in my view, once we get over those and accept the nature of the situation things start to get easier.

Every decision boils down to three things:  1.  What is the probability of an avalanche?  2.  What will happen if one of us is in it?  3.  Is the benefit of crossing this slope worth that risk?

Here's an example.

Chance of triggering - 1 in 1,000.  Worst possible outcome - broken arm.  Benefit - access a long sought-after winter route.

What would you do?  I think many of us would accept those odds and stakes and go ahead.

What about this one?

Chance of triggering - 1 in 30.   Probable outcome - swept off 30m cliff in a size 2 avalanche.  Benefit - climb a route you've done three times before.

No thanks - I'll go and do something else.

Of course the practical side of using this hinges on our ability to estimate the likelihood and size of an avalanche.  Fortunately avalanche forecasts provide a professional assessment of those things so they're a great start but there's a good reason why they're written in the language of probability.  The hazard ratings are full of words like "possible" and "may" and "likely."  We're dealing with uncertainty so nothing there is either a red light or a green one.

If you're then going to go into the field and make further observations on your chosen route and effectively update the forecast as you go then it really means committing to developing the skills to do that.  The good news is these are skills - it's not some magic intuition stuff.  So learn about the snowpack and weather signs to look for.  Understand the significance of dense layers overlying weaker layers.  See what you can and can't find out by digging pits (by the way, it can be quite a lot.)  Find out what kinds of tests can give you more information in different conditions.  At the same time learn to assess terrain.  The great thing with terrain is that it doesn't need judgement because you can measure it.

Putting all of that together gives us a way to more accurately assess probability and consequence from which we can better decide for ourselves, is it worth it?

For anyone whose interest might have been piqued, I highly recommend "Thinking in Bets" by Annie Duke for a crystal-clear take on why we should see more of our decisions in gambling terms.

And thanks DaveHK for starting this thread in the way you did.

edit - spelling

The above would be absolutely fine for decision making if anyone had any idea if a particular slope was 1 in 1000 or 1 in 30. The numbers unfortunately are purely speculative. And 1 in 30 doesn't sound like terrible odds anyway if you don't go on dodgy-looking slopes very often. 

I agree, in so far as I think adrian's use of numerical odds is a bit misleading. Numbers suggest precision, that's not the case but things can still be evaluated. 

Replace the 1 in xx with something meaning "words of estimative probability" and his post makes perfect sense and is a pretty good summary.