PaddleWise by thread

From: Bradford R. Crain <crainb_at_pdx.edu>
subject: [Paddlewise] ACA Accident Report - Critical Judgement II
Date: Sun, 20 Jan 2008 19:42:58 -0800
   According to the ACA Report, Table 13 below, roughly equal
numbers of novice paddler and experienced paddler fatalities
were tabulated in this study, giving the false impression
that novice and experienced paddlers are dying at the same
rate. The counts have not been adjusted for total hours of
risk exposure per paddler, nor has the data been adjusted
for degree of severity of environment in which each paddler
has been operating. Clearly this data should be adjusted for
total risk exposure, just as a cancer patient would be. I find
this table terribly misleading.

Table 13: Canoe and Kayak Fatalities by
Operator Experience 1996-2002
-----------------------------------
|      Hours           | Per Cent |
-----------------------------------
|    <10 hours         |  29%     |
-----------------------------------
|   10-100 hours       |  39%     |
-----------------------------------
|   over 100 hours     |  31%     |
-----------------------------------
Source: USCG Recreational Boating Accident
Report Database (BARD) 1996-2002



-- 
Bradford R. Crain
***************************************************************************
PaddleWise Paddling Mailing List - Any opinions or suggestions expressed
here are solely those of the writer(s). You must assume the entire
responsibility for reliance upon them. All postings copyright the author.
Submissions:     PaddleWise_at_PaddleWise.net
Subscriptions:   PaddleWise-request_at_PaddleWise.net
Website:         http://www.paddlewise.net/
***************************************************************************
From: Dave Kruger <kdruger_at_pacifier.com>
subject: Re: [Paddlewise] ACA Accident Report - Critical Judgment II
Date: Mon, 21 Jan 2008 00:57:15 -0800
Bradford R. Crain wrote:
> According to the ACA Report, Table 13 below, roughly equal numbers of
> novice paddler and experienced paddler fatalities were tabulated in this
> study, giving the false impression that novice and experienced paddlers
> are dying at the same rate. The counts have not been adjusted for total
> hours of risk exposure per paddler, nor has the data been adjusted for
> degree of severity of environment in which each paddler has been
> operating. Clearly this data should be adjusted for total risk exposure,
> just as a cancer patient would be. I find this table terribly
> misleading.

Excellent points, Brad.  Nor has it been adjusted for "paddler skill" in 
the face of difficult/dangerous conditions.  IOW, for a really skilled 
paddler, Class III WW is about as dangerous to that paddler as Class I is 
to a novice.

 > Table 13: Canoe and Kayak Fatalities by Operator Experience 1996-2002
-----------------------------------
|      Hours           | Per Cent |
-----------------------------------
|    <10 hours         |  29%     |
-----------------------------------
|   10-100 hours       |  39%     |
-----------------------------------
|   over 100 hours     |  31%     |
Jeez, Brad, are you surprised they can't do math?  They had two PhD's, an 
MD, and a USCG "Boating Safety Statistician" on board for this report. 
[big, cheesy grin: we all know "PhD" stands for "Piled Higher and Deeper" 
in the hierarchy of "BS" and "MS" ("More of the Same");  BTW, I are one of 
the piled higher and deeper ones, so I should know!]

Their comment about this table _does_ suggest, in a backhanded way, that 
there is a higher than _expected_ rate for the novices:

"Another risk factor for mishap and injury among paddlers is inexperience. 
Information on experience is only available for 323 of the 574
known canoe and kayak fatalities in the USCG database. The majority of 
these fatalities had more than 10 hours of experience (71%) with
just less than a third reporting more than 100 hours of experience (Table 
13)."

I think they are trying to say that, other things being equal, you'd 
_expect_ a much higher rate of fatalities among more experienced paddlers, 
because of greater risk exposure, but the rate for novices is _as_high_ as 
that for experienced paddlers.  Meaning:  if novices were better trained in 
really basic safety practices, their rate should be lesser than the rate 
for folks exposed for longer periods of time to more risky conditions.

Yeah, yeah, I know.  They did not compare the number of fatalities to 
_how_many_ folks were in each group, or to the number of hours of 
exposure/paddler.

A better table would have been one which compared "deaths per paddler-hour 
on the water."  Probably they had no way to get at that figure, except 
conjecture.  Not to mention that there are no experience data for some 40% 
of the 574 known fatalities.

In the end, I think the focus should be on "stupid deaths," in which we 
identify deaths that are easily prevented and get folks to avoid doing 
stupid things like boating drunk.  And, then, accept that in any risky 
activity, some people are going to freaking die!!  Life is short.  As Janis 
Joplin said:  "Get it while you can ..."

-- 
Dave Kruger
Astoria, OR
***************************************************************************
PaddleWise Paddling Mailing List - Any opinions or suggestions expressed
here are solely those of the writer(s). You must assume the entire
responsibility for reliance upon them. All postings copyright the author.
Submissions:     PaddleWise_at_PaddleWise.net
Subscriptions:   PaddleWise-request_at_PaddleWise.net
Website:         http://www.paddlewise.net/
***************************************************************************
From: Bradford R. Crain <crainb_at_pdx.edu>
subject: Re: [Paddlewise] ACA Accident Report - Critical Judgment II
Date: Mon, 21 Jan 2008 08:39:02 -0800
   Not only has "hours spent on the water" not been factored in, but as
you pointed out, there is no mention of total numbers of paddlers  
(dead and alive) in each experience class. Paddlers who did not die  
were "censored" out
of this study. This kind of data falls into the area of Survival Analysis,
a branch of Statistical Science. Survival analysis uses all the information
available from both fatalities and censored survivors, and allows concomitant
variables to be used as well, such as age, gender, etc. A good study of
paddler fatalities should be done by someone well-trained in survival
analysis. This is too specialized an area even for a good general  
statistician.
An analogous study would be examinating survival rates of cancer  
patients, where
all the data is utilized, both from patients who die, and patients who either
drop out of the study, or live beyond termination of the study.
   Another complicating factor to consider: suppose a paddler participates in
kayaking for 50 years, and at age 70 has a heart attack while on the water
and capsizes and dies. Will that count as an experienced kayaker fatality?

Brad

Quoting Dave Kruger <kdruger_at_pacifier.com>:

> Bradford R. Crain wrote:
>> According to the ACA Report, Table 13 below, roughly equal numbers of
>> novice paddler and experienced paddler fatalities were tabulated in this
>> study, giving the false impression that novice and experienced paddlers
>> are dying at the same rate. The counts have not been adjusted for total
>> hours of risk exposure per paddler, nor has the data been adjusted for
>> degree of severity of environment in which each paddler has been
>> operating. Clearly this data should be adjusted for total risk exposure,
>> just as a cancer patient would be. I find this table terribly
>> misleading.
>
> Excellent points, Brad.  Nor has it been adjusted for "paddler skill"
> in the face of difficult/dangerous conditions.  IOW, for a really
> skilled paddler, Class III WW is about as dangerous to that paddler as
> Class I is to a novice.
>
>> Table 13: Canoe and Kayak Fatalities by Operator Experience 1996-2002
> -----------------------------------
> |      Hours           | Per Cent |
> -----------------------------------
> |    <10 hours         |  29%     |
> -----------------------------------
> |   10-100 hours       |  39%     |
> -----------------------------------
> |   over 100 hours     |  31%     |
> Jeez, Brad, are you surprised they can't do math?  They had two PhD's,
> an MD, and a USCG "Boating Safety Statistician" on board for this
> report. [big, cheesy grin: we all know "PhD" stands for "Piled Higher
> and Deeper" in the hierarchy of "BS" and "MS" ("More of the Same");
> BTW, I are one of the piled higher and deeper ones, so I should know!]
>
> Their comment about this table _does_ suggest, in a backhanded way,
> that there is a higher than _expected_ rate for the novices:
>
> "Another risk factor for mishap and injury among paddlers is
> inexperience. Information on experience is only available for 323 of
> the 574
> known canoe and kayak fatalities in the USCG database. The majority of
> these fatalities had more than 10 hours of experience (71%) with
> just less than a third reporting more than 100 hours of experience
> (Table 13)."
>
> I think they are trying to say that, other things being equal, you'd
> _expect_ a much higher rate of fatalities among more experienced
> paddlers, because of greater risk exposure, but the rate for novices is
> _as_high_ as that for experienced paddlers.  Meaning:  if novices were
> better trained in really basic safety practices, their rate should be
> lesser than the rate for folks exposed for longer periods of time to
> more risky conditions.
>
> Yeah, yeah, I know.  They did not compare the number of fatalities to
> _how_many_ folks were in each group, or to the number of hours of
> exposure/paddler.
>
> A better table would have been one which compared "deaths per
> paddler-hour on the water."  Probably they had no way to get at that
> figure, except conjecture.  Not to mention that there are no experience
> data for some 40% of the 574 known fatalities.
>
> In the end, I think the focus should be on "stupid deaths," in which we
> identify deaths that are easily prevented and get folks to avoid doing
> stupid things like boating drunk.  And, then, accept that in any risky
> activity, some people are going to freaking die!!  Life is short.  As
> Janis Joplin said:  "Get it while you can ..."
>
> -- 
> Dave Kruger
> Astoria, OR
***************************************************************************
PaddleWise Paddling Mailing List - Any opinions or suggestions expressed
here are solely those of the writer(s). You must assume the entire
responsibility for reliance upon them. All postings copyright the author.
Submissions:     PaddleWise_at_PaddleWise.net
Subscriptions:   PaddleWise-request_at_PaddleWise.net
Website:         http://www.paddlewise.net/
***************************************************************************
From: Rich Kulawiec <rsk_at_rockandwater.net>
subject: Re: [Paddlewise] ACA Accident Report - Critical Judgement II
Date: Mon, 21 Jan 2008 21:02:14 -0500
On Sun, Jan 20, 2008 at 07:42:58PM -0800, Bradford R. Crain wrote:
>   According to the ACA Report, Table 13 below, roughly equal
> numbers of novice paddler and experienced paddler fatalities
> were tabulated in this study, giving the false impression
> that novice and experienced paddlers are dying at the same
> rate. The counts have not been adjusted for total hours of
> risk exposure per paddler, nor has the data been adjusted
> for degree of severity of environment in which each paddler
> has been operating. Clearly this data should be adjusted for
> total risk exposure, just as a cancer patient would be. I find
> this table terribly misleading.

This is a superb point.  And I'd include in that "severity of
environment" factors such as:

	- whitewater class
	- whitewater type (continuous vs. pool-and-drop)
	- weather conditions
	- remoteness
	- river levels as compared to "normal"
	- etc.

Various attempts at quantifying factors like those have been made over
the years.  A notable example may be found in Monte Smith's "Southeastern
Whitewater" where he explains his "TRIP" scale and uses it to assess
several dozen well-known runs throughout the southeast US.  Of course
his scheme, like any other, is largely empirical -- but based on my
own first-hand experience on a number of those runs, it's at least
in the ballpark.  Most of its utility lies not in the absolute values
of the numbers -- which are, after all, arbitrary -- but in the *differences*
of those values for different runs.  Those differences enable paddlers
familiar with run A but not run B to reach conclusions such as "run B
is about as hard as run A" or "B is markedly more difficult than A"
with reasonably high confidence that those conclusions are accurate.

	[ Briefly, here's how the TRIP scale works.  Nine numbers
	that assess stream difficulty are (in some cases) measured
	or (in some cases) assigned.  Those numbers are then weighted
	and combined to produce a single TRIP number.  The nine
	factors and their weights are:

	23%	difficulty of rapids
	20%	volume X gradient interaction
	17% 	average gradient
	10%	streambed morphology
	7%	continuousness of rapids
	7%	maximum gradient in any one mile
	6%	total gradient
	6%	inaccessbility
	4%	reputation

	Some folks have wondered why that last one is in there --
	I think it's reasonable.  Paddler psychology *is* a factor:
	even though there are sections of the Upper Meadow that
	are arguably more difficult (especially above 1000 CFS)
	than sections of the Lower Gauley, paddlers on the latter
	have in their heads "I'm on the $#*&! Gauley" and that
	does make a difference.

	Note that he doesn't factor in river level or weather:
	his assessments are made at nomimal water levels (which he
	specifies) and he does include comments about how his
	assessments change under high-water conditions.  And I think
	anyone using his book is likely well aware that paddling
	the New River Gorge on a sunny July day is a profoundly
	different experience than running it in a March snowstorm
	even if water levels are identical. ]

I'm not claiming that Smith's methodology is the best; I just think
it's a credible attempt.  And something roughly along those same lines
needs to factored in to studies like this in order to (as Brad said)
adjust it for total risk exposure.

And I'd also add this factor: familiarity.  I've paddled one local
stream several thousand times (I trained there nearly every day for years)
under every possible condition, including record low and record high.
It's probably "safer" for me than some easier streams, because I know
every rock, every wave, every eddy and exactly what they'll do under
any condition.  The same can probably be said of people who run the
Ocoee or the Clackamas or any other stream all the time.  Whereas if
any of us were on the others' "home river" for the first or second
or Nth time, we'd probably incur higher risk.

---Rsk
***************************************************************************
PaddleWise Paddling Mailing List - Any opinions or suggestions expressed
here are solely those of the writer(s). You must assume the entire
responsibility for reliance upon them. All postings copyright the author.
Submissions:     PaddleWise_at_PaddleWise.net
Subscriptions:   PaddleWise-request_at_PaddleWise.net
Website:         http://www.paddlewise.net/
***************************************************************************

This archive was generated by hypermail 2.4.0 : Thu Aug 21 2025 - 16:33:48 PDT