Topic Options
Rate This Topic
#104585 - Wed Jan 26 2005 08:16 PM evaluating statistics
db le p Offline
SJ'er with 5000+ posts

Registered: Tue Sep 24 2002
Posts: 6723
Loc: Germany
NOTE: Judging by the activity in another thread in this new avvy forum, people may accuse me of stirring trouble. Please believe me that I am not.

Having said that, I swing right into a blunt question: Is it possible that relatively very few skiers and boarders die from avalanches? In North America from 1996 - 2002 (7 seasons):

http://www.avalanche.org/~uac/graf_actdet.html

38% of 146 deaths = 55 people = 8 people per season (average). Possibly not a shocking number. Another useful metric would be of the people that died, how many died in the same av?The greater the 'single event' concentartion, the lesser the individual event risk.

Most important: I really need to see the cumulative number of skiers+boarders in the BC over these 7 seasons.

If 50 people hit the BC on one day, then that counts as 50 BC days. If that happens 30 days per season then that is 1,500 BC days. Over 7 seasons we have a hypothetical 10,500 BC days. In 7 seasons there were 10,500 individual instances of BC risk. Only 55 people died.

(to put the above hypothetical estimate in anopther way yet arrive at the same number: 50 individual people head into the BC 30 times each per season for 7 seasons = 10,500)

A much more conservative and massively unrealistic approach : Lets assume each season has 30 days when someone ventures into the BC. On those days only 1 person heads for the back country. Over 7 season that is 210 BC days and that does not take into account the number of people that enter the back country on those days. In 7 seasons there were 210 individual instances of BC risk. 55 people died.

I would like to hear estimates of the following inputs:

number of BC days per season = X
number of people in the BC on each BC day = Y
therefore number of BC 'events' per season = X*Y

How about:

X = 40 (10 prime BC days per month over a 4 month season)
Y = 100 (skiers and boarders in all North America)
X*Y = 4,000 BC 'events' per season.

In 2002 there were 35 deaths of which 18 were snow mobilers. 35 less 18 = 17 skiers and boarders killed in 2002 by avalanche (over estimate as it includes other recreationalists and av professionals).

17 deaths in 4000 BC events is a small number. Very small.

(souce: http://www.avalanche.org/~uac/graf_usava_year.html less http://www.avalanche.org/~uac/graf_snomo_year.html)

I am the first to point out that the risk may seem small but the down side is rather binary: you lose, you die.

This thread is not a reflection of my attitude towards avalanche risk nor the education surrounding it (although I will take route selection training over beacon/pole/shovel training any day). I will not argue with comments that one death is too many, but I also agree with rational reasoning that people die when they do physical things. I remember hearing about the 'death budget' component of large Army training exercises in which I used to take part. A two week training exercise involving 4000 people was expected to result in a certain number of deaths. It was just a given.

You are right if you claim that I am inexperienced and just basing these comments on written statistics (Bruce Temper's "Staying Alive in Avalanche Terrain" + internet sources) rather than time in the field.

For what it is worth, I find the imperfect science of av prediction and the detailed algorithm of risk determination to be fascinating. It is exactly the type of thinking that my brain enjoys. If I had my time again I would take an outdoors career role in industry which required av risk assessment rather than my current career, finance (where I research and develop new and imperfect 'risk assessments' for my organisation). Having said that, I hope that I have not got any of my above numbers wrong ;\)

Top
#104586 - Wed Jan 26 2005 10:08 PM Re: evaluating statistics
montoya Offline
SJ'er with 1000+ posts

Registered: Tue Jan 29 2002
Posts: 1356
Loc: yuzawa--tokyo
There are some interesting figures here, although it is for France. Would love to get my hands on avie data for Japan someday:

http://mapage.noos.fr/pistehors/images/avalanche/analysis-2004.pdf

pg.14

"It would be interesting to calculate the avalanche fatality rate for skier/days (outings). French ski resorts record about 50 million skier/days per year. Comparing extensive figures for on and off piste interventions by the rescue services, about 2.5% skier/days are spent off-piste. With 11 off-piste deaths last season (the 13 year average is 8.5), this gives an approximate fatality rate of 1 per 96,000 skier/days for off-piste skiing. If the rate is similar for backcountry accidents, this would imply that at present only about 2% of outings are recorded in the SkiRando database. The figures in this paragraph are presented more as an academic exercise and should be treated with caution."

Top
#104587 - Thu Jan 27 2005 10:22 AM Re: evaluating statistics
Toque Offline
SJ'er with 7500+ posts

Registered: Fri May 23 2003
Posts: 7985
Loc: Northwest BC
34 you have think about it this way.

Do you want to ski/board that line so much that you are willing to risk your life doing so?

If yes then go for it.

The odds are on your side that you will live to ride the next one. But eventually it will catch up to you.
It is usually the most experienced people that die in slides. But that is because they are out there day in and day out and eventually it catches up to them.

Myself I am not willing to die for a ski line. But then have I skied some off piste lines where if it slid I would have died. Hell yes.
Not sure what that means.

Top
#104588 - Thu Jan 27 2005 10:27 AM Re: evaluating statistics
HighlyTrainedExNovaTeacher Offline
SJ'er with 200+ posts

Registered: Thu Dec 23 2004
Posts: 203
Loc: Tokyo
Sounds like it might mean you're a bit confused dude ;\)

Top
#104589 - Thu Jan 27 2005 11:25 AM Re: evaluating statistics
Mr Wiggles Online   content
SJ'er with 2000+ posts

Registered: Fri Jul 06 2001
Posts: 2478
Loc: Ye olde Hakuba
Lots of bc activity is going to be low risk because it has factored in avalanche risk. Traditionally the Japanese yama ski folk start hitting the big stuff in spring. What people do in April is no measure of how safe it may be in February one day after a storm.
_________________________
"The toe is the Achilles' heel of the foot."

Top
#104590 - Thu Jan 27 2005 12:03 PM Re: evaluating statistics
Creek Boy Offline
SJ'er with 10000+ posts

Registered: Wed Jul 17 2002
Posts: 11140
Loc: is everything
Damn, Id hate to see what you think about/come up with after smokin a big fatty
_________________________
what did you poach today?

Top
#104591 - Thu Jan 27 2005 07:04 PM Re: evaluating statistics
db le p Offline
SJ'er with 5000+ posts

Registered: Tue Sep 24 2002
Posts: 6723
Loc: Germany
At least I didn't start a fight \:\)

Montoya - good link and quote.

Toque - Like I said, although the odds are in your favour, the outcome is binary: live or die. There are many sports that have this low probability terminal pay-off profile. I tend to avoid them myself.

MrW - I am of course speaking from a position of ignorance, but yes, I think avalanche risk has actually been factored into a lot of BC activity to a reasonable degree. It reinforces the fact that if you measure and manage the risk your chances of death are very very very very low. The evidence is clearly in the numbers. Not many people die from avalanche as they avoid the risk in the first place (not because they dig each other out after a beacon and probe search). It is also a function of the proposed fact that the snowpack is stable 95% of the time. Skiers and boarders seem to come from a position of respect for the forces of nature: they tend to naturally avoid many high risk avalanche situations- an assessment that they make based of common sense and avalanche education. Avalanche education needs to step up its efforts now that the BC is being accessed by weekenders with only a few seasons experience (people like me). As reckless weekend adventures after powder become more popular, which they are, a disproportionately higher number of deaths will occur. Evidence for this is in the high proportion of incidents involving highmarking snowmobilers. The problem in the future will be if this reckless snowmobile attitude is replicated in those skiers/boarders new to the BC environment on better technology (rando and split board gear is opening the BC door for many). The chances are high (that this reckless mindset will be replicated in new ski/board BC participants) and targeted avalanche risk training needs to be put in place. I sheepishly admit that I never once in the past gave one sensible thought to the avalanche risk in my Japanese rope ducking activities. I now shudder with embarrassment, but realise that nothing happened as the odds were well in my favour.... until they aren't and I die: I don't like that speculative profile at all and will be changing my behaviour as a result.

Yama - My mind works at a disturbingly unnatural rate of abstraction on a calm day, as such I avoid smokin a big fatty 100% of the time. Smoking just adds fuel to the already complicated fire ;\)

Top
#104592 - Sat Jan 29 2005 09:54 PM Re: evaluating statistics
SerreChe Online   content
SJ'er with 1000+ posts

Registered: Sun Dec 03 2000
Posts: 1687
Loc: Prev 6° 37' 2'' E 44° 53' 12''...
What kind of call spread can you put on to hedge that digital risk?

Top
#104593 - Mon Jan 31 2005 10:02 PM Re: evaluating statistics
db le p Offline
SJ'er with 5000+ posts

Registered: Tue Sep 24 2002
Posts: 6723
Loc: Germany
Don't you have a brother in the swap market? He could help you ;\)

I have actually been thinking about the 'downside profile' (ie, the shape of the outcome if you lose). At first I thought it was like a step: live or die, no inbetween was important. But now I think that there are important shades of grey. A rough ranking based on my view on life.

- Friend is caught, easily rescued, no injury (a scare)
- I am caught, easily rescued, no injury
- Friend is caught, rescued but injured
- I am caught, easily rescued but injured
- Friend is caught, rescued but severely injured
- My wife is caught, easily rescued, no injury (note: I dont have a wife)
- My wife is caught, easily rescued but injured
- I am caught, easily rescued but severely injured
- Friend is caught, killed
- My wife is caught, easily rescued but severely injured
- I am caught, killed
- My wife is caught, killed (an utter nightmare).

The chances of an individual skier/boarder dying in an avalanche is quite low as I discussed above. However the chances of any number of nasty outcome could be quite a bit higher. Even the 'least bad' of my negative outcomes is not what I would ever want, ever. It is my plan to equip myself so that I can maximise <30 degree slopes as often as possible. If I ride >30 degree slopes more than a few times in a season then I am probably pushing my luck: the unlikely negative pay-off just is not worth the risk. The only possible exceptions would be possibly >45 degree glacier/couloir riding in spring or if I was with a very experienced guide who lives in the snow pack I am riding. Even then I am not at all comfortable with the concept of paying someone and then consigning them the right to risk my life.

(note, just for clarification: the list shows a set of negative outcomes, with the worst being at the bottom. It may seem a laborious thing to do, but if you are out taking risk and have not contemplated all the likely outcomes and considered what effect they might have on you then you are not taking the risk seriously when making your risk assessment. Ask yourself if you want to take out your beacon to search for your wife/brother/son.... even if they are quickly found with no injury. As I have already said, that is not a situation I ever want to be in).

Top
#104594 - Tue Feb 01 2005 11:40 PM Re: evaluating statistics
SerreChe Online   content
SJ'er with 1000+ posts

Registered: Sun Dec 03 2000
Posts: 1687
Loc: Prev 6° 37' 2'' E 44° 53' 12''...
Got to beware with these swap boys always dodgy ;\) Do I know u 34? Do not think so since u r from Rondon. Anyway, looks like to me you need to fire up the monte carlo on that puppy (do not forget the positive outcomes of pure ecstasy in there).

Top


Moderator:  SJ#2, SJ#3, SJ#4, SJ#5, SJBot, SJForums 
SnowJapan Chat Box
The SnowJapan Chat Box is currently closed.
Who's Online
26 Registered (art vandelay, boardbaka, DG, Gary, Go Native, golf74, grazza, jetpilot000, Mantas, me jane, MikePow, Mr Wiggles, Mudguts, Paris88, ratherb@thesnow, Richardkorea, Sciclone, SerreChe, shadowtec, snowhunter, stemik, tdc, Tubby Beaver, XilR8, 2 invisible), 85 visible guests and 19 others.
Key: Admin, Global Mod, Mod
Forum Stats
7220 Members
11 Forums
16172 Topics
295581 Posts
182 Average daily posts (last week)
129 Posts in the last 24 hrs

Max Online: 2075 @ Fri Apr 04 2008 09:23 AM
Newest Members
dools, CigaretteSmokingMan, jetpilot000, Jessta, Richardkorea, MuminHongKong, HongKong
7220 Registered Members
SnowJapan News & Announcements
RESORT SPOTLIGHT - Furano (Furano, Hokkaido)
by SJ#4
Yesterday at 08:11 PM
SHIGA KOGEN ACCOMMODATION - Bersarooms (Shiga Kogen, Nagano)
by SJ#4
Yesterday at 05:04 PM
RESORT SPOTLIGHT - Dynaland (Takasu, Gifu Prefecture)
by SJ#4
Yesterday at 04:02 PM
About weather forecasts & snow depth data
by SJ#4
Thu Nov 20 2008 08:38 AM
HAKUBA ACCOMMODATION - White Horse House (Hakuba, Nagano)
by SJ#4
Fri Nov 14 2008 11:20 AM
Daily Reports
22nd November 2008 - daily Japan snow & weather reports
by SJ#4
Today at 08:36 AM
21st November 2008 - daily Japan snow & weather reports
by SJ#4
Yesterday at 09:12 AM
SnowJapan Information
SnowJapan.com
Interview about SnowJapan
About SnowJapan
Site Map
News & Announcements
Contact Us
Forums Registration


SNOWJAPAN.COM   :   SNOWJAPAN.TV
the japan winter sports guide & community

JAPAN SKI & SNOWBOARD RESORTS
Japan Snow Resorts Index : Resorts by Prefecture : Resort Spotlights : Ski Town Maps : Resort Reviews : Resort Photos : Search Resorts
JAPAN SKI ACCOMMODATION & PLACES TO STAY
Places to Stay Index (All Regions) : Hokkaido : Nagano : Niigata : Gunma : Iwate : Yamagata : Fukushima
JAPAN SNOW & WEATHER DAILY REPORTING
Daily Reporting Index : Niseko Now : Furano Now : Hakuba Now : Nozawa Now : Shiga Kogen Now : Yuzawa Now : Naeba Now : Minakami Now : Now Calendar : Resort Webcams
JAPAN SKI TOWN GUIDES
Town Guides Index : Niseko Town Guide : Furano Town Guide : Hakuba Town Guide : Nozawa Town Guide : Shiga Kogen Town Guide : Yuzawa & Naeba Town Guide : Minakami Town Guide
GUIDES & INFORMATION
Travel Info : General Info : Services : Lessons in English : Tours & Services : Jobs Available : Features & Interviews : Photo Galleries : Reader Photo Galleries

About SnowJapan : Contact Us : Site Map : Search : Start Here!
SnowJapan © 2008
www.snowjapanforums.com