[pct-l] Probability of Disaster

Wed Sep 1 21:58:03 CDT 2010

Steel-eye, 

Part of your argument is valid and part is flawed, namely the "sequential" probability analysis.  I used to teach probablity theory in college, but I've forgotten a lot of the arcane details.  However, multiplying P(a) times P(b) times ... to produce the P(of all occuring at once) is valid if and only if a, b, c, ... are independent variables.  In your senario, the five occurences are not independent but closely, albeit assumptively , related.   

Your argument has intuitive validity but the assumptions and methodology are not statistically sound.  

Mango 
----- Original Message ----- 
From: "CHUCK CHELIN" <steeleye at wildblue.net> 
To: "PCT listserve" <pct-l at backcountry.net> 
Sent: Wednesday, September 1, 2010 7:00:28 PM 
Subject: [pct-l] Probability of Disaster 

Good afternoon, all, 

I represent the ultra-lite faction when discussing gear.  For a 
long-distance hike I prefer to keep my base weight around 8 lbs, and seldom 
conditionally stretch that to 10-12 lbs.  I’ll admit one of my basic credos 
is, “When in doubt, leave it out.”  That doesn’t leave a lot of margin for 
“difficulty”, but having successfully survived wilderness situations for 
about 50 years I feel comfortable with that approach. 

Sir Winston Churchill once said, “Why, you may take the most gallant sailor, 
the most intrepid airman or the most audacious soldier, put them at a table 
together- what do you get?  The sum of all fears.”  Author, Tom Clancy, like 
that comment so much that he borrowed part of it to use as the title of one 
of his books. 

PCT-L is much like that meeting table:  One expert claims a 4-season tent is 
necessary.  Another says lots of layers of clothes is best.  One claims 
communication is the key with a phone, a two-way radio, and a SPOT.  Another 
advocates having full-on alpine climbing gear.  Occasionally someone feels 
they need to carry a gun to foil the large, carnivorous wildlife.  Someone 
else simply must have a serious, expedition-capable first-aid kit.  It goes 
on and on: All of their arguments seem to have merit but if we take the sum 
of their recommendations to heart we end up carrying a huge load. 

Those who advocate carrying lots of “what if” gear usually claim they want a 
greater margin in the event of a problem.  They say, “What if … (something 
bad)… happens?  I would certainly … (suffer somehow) … with no-one around to 
help.  If I just pack … (an additional item) … I could use it to save my 
life.” 

That all sounds very sensible as we play these “what if” games in our mind, 
but let’s parse that statement and see what it really means.  That statement 
presupposes: 

1)       Something bad could happen, and, 

2)      The result could be a disaster, and, 

3)      There are no options to consider, and, 

4)      An additional piece(s) of gear is the answer, and, 

5)      I can use that gear to resolve the disaster. 

If any of those suppositions are untrue – meaning having a near-0% 
probability of occurring – the entire scenario is either not going to happen 
or the results will be significantly mitigated.  If all five of those 
suppositions are possible, the probability of each actually happening is 
sequentially multiplied.  That’s why it’s called sequential conditional 
logic. 

For example, bad things may be possible but they don’t often happen – except 
in our minds.  What’s the chance of our being caught in a freak snowstorm, 
or being stranded on a high-angle snowfield, or encountering a really deep, 
dangerous ford, etc.?  Pick a number – let’s say it’s 20%. 

Next, what’s the chance of that event – if it happens -- really being a 
potential disaster?  I don’t mean the discomfort of being soaked to the 
skin; or the inconvenience of a cold, uncomfortable night; I mean something 
that genuinely threatens life and limb?  Again, pick a number -- let’s say 
it’s 10%. 

Also, what are the chances that there are no options available to mitigate 
the potentially disastrous possibility?  Is there no-one around to lend a 
hand?  Is there no re-route that can be taken?  Can one not bail out on a 
convenient side trail?  Is it not possible to just continue to hike – even 
night-hike -- to stay warm?  I submit there are almost always options to 
avoid disaster, but let’s say there’s a 20% chance that no viable options 
are available. 

Next, what’s the chance that an extra piece(s) of gear will really save the 
day?  It can happen:  Maybe a nice 4-season tent would help.  Maybe a 
generous array of alpine climbing gear would get us off the hook.  Maybe a 
GPS would help us find our way back to the trail. Maybe a really full-on set 
of storm gear with extra clothing would save the day, etc.  Gear won’t get 
us out all troubles, but I’ll say there’s a 50% chance it will. 

Finally, what’s the probability that we actually know how to effectively 
employ that extra gear?  A tent we can pitch; clothes we can put on, etc, 
but many – or most -- hikers can’t use climbing gear; most can’t use GPS 
effectively; and the contents of a big “survival kit” or an expedition 
first-aid kit are largely a mystery.  A gun?  Very likely the average hiker 
couldn’t hit a circus tent with it at any range over 10 feet.  I guess the 
odds of a hiker being actually able to use specialized extra gear to be 50% 
at best. 

So here’s how it stacks up:  The likelihood, 

… of an event happening, 20% 

… of it being potential disaster,  10% 

… of few/no options or help being available,  20% 

… of gear being the substantial answer, 50% 

… of expertise being available to employ gear, 50% 

Now comes the “sequential” part:  Multiply 0.20 x 0.10 x 0.20 x 0.50 x 0.50 
and see that the result of that conditional string equals 0.001 which is 
0.1%.  That’s a one tenth of one percent likelihood of all that happening.  I, 
for one, am not going to carry very much extra gear against such a 
probability.  Don’t quarrel with my numbers; plug in your own. 

Your results may vary – as will your pack weight. 

“Tools are deceiving.  They have a way of making you think you can use them, 
as you study them and imagine yourself owing them...  There is no name for 
this quality that tools -- have, but it’s undeniable.” 

-- Geoffrey Norman, *Outside*, Oct. 1986 

Steel-Eye 

Hiking the Pct since before it was the PCT – 1965 

http://www.trailjournals.com/steel-eye 

http://www.trailjournals.com/SteelEye09 
_______________________________________________ 
Pct-L mailing list 
Pct-L at backcountry.net 
To unsubcribe, or change options visit: 
http://mailman.backcountry.net/mailman/listinfo/pct-l 

List Archives: 
http://mailman.backcountry.net/pipermail/pct-l/ 