Invisible Cyclists in Solstice Dark

  • Thread starter Elisa Francesca Roselli
  • Start date



SMS wrote:
>
> However there are at least a few of us that believe that the junk
> science, logical fallacies, and mis-use of statistics, do not do the
> anti-compulsion cause any good.


Such hand-waving!

Why not give us specifics? Why are you so incapable of giving us
concrete examples? One begins to suspect you have nothing concrete to
discuss - just your own half-baked opinions, yet again!

> Such arguments may convince people that
> lack critical thinking skills, but those are not the people that really
> matter when you're either fighting to stop an MHL, or trying to get one
> repealed.


Said by a person who has never successfully fought a MHL.

I've used the points I've made here in (IIRC) three successful attempts
to defeat such a law. In one case, I spoke directly to one of the laws
sponsors in a one-on-one conversation. He was very attentive, and
apparently impressed. I noted the next incarnation of that bill did
not list him as a co-sponsor.

Based on my experience, Steven, I'd say you have no idea how to fight a
helmet law. You really should stop pretending you're an expert.

- Frank Krygowski
 
On 11 Jan 2006 11:04:28 -0800, [email protected] said in
<[email protected]>:

>I've used the points I've made here in (IIRC) three successful attempts
>to defeat such a law. In one case, I spoke directly to one of the laws
>sponsors in a one-on-one conversation. He was very attentive, and
>apparently impressed. I noted the next incarnation of that bill did
>not list him as a co-sponsor.


Same here: one actual bill, two amendments and another extant law now
being challenged, all thanks to precisely the arguments Scharf states
can't possibly work. But then, for Scharf to assert from ignorance is
nothing new :)

Guy
--
http://www.chapmancentral.co.uk

"To every complex problem there is a solution which is
simple, neat and wrong" - HL Mencken
 
Just zis Guy, you know? wrote:
> On 11 Jan 2006 11:04:28 -0800, [email protected] said in
> <[email protected]>:
>
>> I've used the points I've made here in (IIRC) three successful attempts
>> to defeat such a law. In one case, I spoke directly to one of the laws
>> sponsors in a one-on-one conversation. He was very attentive, and
>> apparently impressed. I noted the next incarnation of that bill did
>> not list him as a co-sponsor.

>
> Same here: one actual bill, two amendments and another extant law now
> being challenged, all thanks to precisely the arguments Scharf states
> can't possibly work. But then, for Scharf to assert from ignorance is
> nothing new :)
>


I'm still waiting for him to assert in which year mandatory helmet laws
were introduced on the graphs I linked to. Its funny how he continually
stays silent on taking that challenge. I wonder why that would be?


--
Tony

"The best way I know of to win an argument is to start by being in the
right."
- Lord Hailsham
 
On 11 Jan 2006 07:57:26 -0800, "gds" <[email protected]> said in
<[email protected]>:

>> >As the author of _Econometric Modeling as Junk Science_ wrote: "How much
>> >time should researchers spend replicating and criticizing studies using
>> >methods that have repeatedly failed?


>> And since when did that ever stop a helmeteer form trying, and
>> failing, one more time, to repeat the 85% in the Seattle study? Oh,
>> but wait, the studies which support your prejudices are accepted
>> without question, aren't they?


>No!!!
>And you just don't listen. There is a whole host of folks who are
>criticiszing the studies on both sides. You constantly keep throwing
>out red herrings with great displays of frustration. However, you
>continually mis quote and mis represent what others say.


Here is the problem: pro-helmet studies are deeply and fatally flawed.
I don't think you are disputing that. So: the proponents of the
intervention have failed to prove their case. No problem.

You seem to want the sceptics to prove their case. That is not
possible, because it requires proof of a negative. All we know is
that there are pro-helmet studies which all suffer from self-selection
and confounding, and sceptical studies, almost all of which are
enormously larger, and which have a variety of different flaws.

So in the end it comes back to being a matter of personal choice, a
situation with which I have always been perfectly happy.

And if everyone left it at that instead of trying to force their
opinions on others, that would be an end of it, but the True
Believers, denying both the existence of conflicting evidence and the
flaws in their own, want to force their opinion on others, by
legislation for preference since we stubbornly refuse to accept their
view of how dangerous cycling is, and how effective helmets are. We
are not allowed to make up our own minds unless we agree with them.

>Here it is Guy, Frank, Tony, whomever. I am not convinced helmets are
>good. bad, or indifferent. Why, becasue I have seen no studies that
>convince me of anything. There are lots of studies that "suggest"
>conclusions that need to be properly studied.I do not defend bad
>studies on any side of the issue.


Perfect. You and I are both in complete agreement. I think the
sceptical studies are somewhat better, in that different studies with
different flaws all agree, but in the end the evidence is conflicting
and unsatisfactory on all sides, and only a fool denies it. The sort
of fool who asserts that "all" the research agrees (on anything!)

Guy
--
http://www.chapmancentral.co.uk

"To every complex problem there is a solution which is
simple, neat and wrong" - HL Mencken
 
On Wed, 11 Jan 2006 10:27:43 -0800, SMS <[email protected]>
said in <[email protected]>:

>Especially for the Thompson RS, Rivara, FP, Thompson
>DC study. And what's rather amusing about this is that it's not the
>study itself that is attacked, because the study, including the
>revisions, simply report their results of the research (and the
>conclusion should be viewed in that context).


A valid criticism since this is far and away the most widely cited
figure for efficacy.

>It's all the various
>journals, politicians, etc. that quote the results of that study without
>any disclaimer regarding the study's limitations, and the fact that the
>study made no attempt to account for other variables.


A valid criticism since many studies start form the premise that the
conclusions of TR&T are correct.

Guy
--
http://www.chapmancentral.co.uk

"To every complex problem there is a solution which is
simple, neat and wrong" - HL Mencken
 
SMS wrote:
> peter wrote:
>
> > My situation is very similar. I was originally of the opinion that
> > cycling helmets would be of significant value and had been wearing one
> > since the introduction of the Bell Biker model around 1975. But after
> > seeing some of the helmet discussions I looked into the research in
> > more detail, particularly the widely cited Thompson/Rivara case control
> > study, and changed my mind on the subject.

>
> Your mistake here is to read more into the Seattle study than the study
> itself reported.


Talk about reading more into something than is actually written - your
statement certainly does that.

> This is understandable if you look only at the cites of
> the Seattle study, which invariably quote the 85% or 88% numbers without
> any explanation of how the study was conducted, or its limitations.
>

This is of course the exact opposite of what I stated above; i.e. that
I was originally of the opinion that helmets did have significant value
which was based on a superficial reading of study results, including
the 85/89% figures frequently cited, but changed my mind upon reading
the Thompson/Rivara/Thompson study in detail and seeing the fundamental
flaws in their research.

> In any case, I doubt that you'll find anyone in this thread, or any of
> the other helmet threads, that are in favor of compulsory helmet laws.
> However there are at least a few of us that believe that the junk
> science, logical fallacies, and mis-use of statistics, do not do the
> anti-compulsion cause any good.


But it's the advocates of MHLs that are citing all the truly junk
science case control studies like the ones by Thompson/Rivara. These
use self-selected population samples and erroniously base their
conclusions on the *assumption* that the only explanation for different
injuries between the groups is their choice of head covering.
My observations of cyclists shows that helmet and non-helmet wearing
groups differ in many other respects. The helmet wearers are far more
likely to be riding in groups and out for training/exercise, they
generally follow vehicular traffic patterns and choose routes largely
based on their suitability for cycling. OTOH, far more of the
non-helmet users are transportational cyclists riding alone, frequently
using a mix of vehicular and pedestrian modes, and on routes based on
the location of end destinations (work/shops/etc.). Given how
different the circumstances and cycling styles of these groups are I'd
expect them to have very different injury patterns even if they were
all wearing the same head covering.

> Such arguments may convince people that
> lack critical thinking skills, but those are not the people that really
> matter when you're either fighting to stop an MHL, or trying to get one
> repealed.


On the contrary, I've found that pointing out the fundamental flaws of
case control studies and how their conclusions are further brought into
question by contradictory findings in the whole population studies in
Australia and New Zealand is quite effective in convincing those with
the best critical thinking skills. From reports here it's also been
apparent that Avery, Frank, Guy, and others have successfully used such
discussions in defeating a variety of proposed MHLs.
 
peter wrote:

> But it's the advocates of MHLs that are citing all the truly junk
> science case control studies like the ones by Thompson/Rivara. These
> use self-selected population samples and erroniously base their
> conclusions on the *assumption* that the only explanation for different
> injuries between the groups is their choice of head covering.


No they don't. The issue with the Seattle study is that was limited to
people that sought treatment, the results were accurate for that small
subset of people. Same with the FARS data, the data is accurate, the
statistics generated from the data are accurate, but again, it's data
that is limited to the subset of people that were in reported accidents.
If these studies are quoted, then they should be quoted in a way that
makes clear that the results are based on a certain sample, but they
rarely are.

It's the politicians that are turning these ER and accident studies into
something more than they really are. This contrasts with the various
studies that have been mentioned here regarding countries with MHLs,
where the wild logical fallacies, false assumptions, and statistical
methods are hopeless. You'll see these "studies" endlessly referenced,
and superficially they look impressive, but only to people that are not
trained in statistical methods.

> My observations of cyclists shows that helmet and non-helmet wearing
> groups differ in many other respects. The helmet wearers are far more
> likely to be riding in groups and out for training/exercise, they
> generally follow vehicular traffic patterns and choose routes largely
> based on their suitability for cycling. OTOH, far more of the
> non-helmet users are transportational cyclists riding alone, frequently
> using a mix of vehicular and pedestrian modes, and on routes based on
> the location of end destinations (work/shops/etc.). Given how
> different the circumstances and cycling styles of these groups are I'd
> expect them to have very different injury patterns even if they were
> all wearing the same head covering.


I wouldn't dispute that, though there is of course no evidence either
way. The number of injuries and fatalities is so small to begin with,
that it would require a larger sample than normal to prove this either way.

>> Such arguments may convince people that
>> lack critical thinking skills, but those are not the people that really
>> matter when you're either fighting to stop an MHL, or trying to get one
>> repealed.

>
> On the contrary, I've found that pointing out the fundamental flaws of
> case control studies and how their conclusions are further brought into
> question by contradictory findings in the whole population studies in
> Australia and New Zealand is quite effective in convincing those with
> the best critical thinking skills.


It's these whole population studies that are the junk science. If they
convince anyone, then those people lack the skill to recognize junk science.

> From reports here it's also been
> apparent that Avery, Frank, Guy, and others have successfully used such
> discussions in defeating a variety of proposed MHLs.


Only if you believe that their statements are more accurate than the
junk science that they routinely post.
 
On Wed, 11 Jan 2006 13:20:32 -0800, SMS <[email protected]>
said in <[email protected]>:

>No they don't. The issue with the Seattle study is that was limited to
>people that sought treatment, the results were accurate for that small
>subset of people.


For values of "accurate" which include comparing completely dissimilar
groups of cyclists and attributing all the difference in injuries to
helmet use. And then adding ten percent to the efficacy figure to
compensate for it.

Guy
--
http://www.chapmancentral.co.uk

"To every complex problem there is a solution which is
simple, neat and wrong" - HL Mencken
 
SMS wrote:
>
> It's the politicians that are turning these ER and accident studies into
> something more than they really are.


No, it's the helmet lobbyists that are doing this - funded in part by
the helmet companies. The politicians are in the audience, along with
the general public. They are not at the podium.

> This contrasts with the various
> studies that have been mentioned here regarding countries with MHLs,
> where the wild logical fallacies, false assumptions, and statistical
> methods are hopeless. You'll see these "studies" endlessly referenced,
> and superficially they look impressive, but only to people that are not
> trained in statistical methods.


Oh, please, dear expert in statistical methods, DO tell us
_specifically_ what is wrong with _specific_ studies, won't you? To
date, all you've done is waved your hands and whined "junk science,
junk science."

If you can't cite specifics, you look like a fool entranced by the
sound of his own typing.

> It's these whole population studies that are the junk science. If they
> convince anyone, then those people lack the skill to recognize junk science.


More hand-waving. Obviously, you're afraid of giving specifics,
because you're afraid of being proven wrong.

> Only if you believe that their statements are more accurate than the
> junk science that they routinely post.


And still no specifics.

- Frank Krygowski
 
SMS wrote:
> You'll see these "studies" endlessly referenced,
> and superficially they look impressive, but only to people that are not
> trained in statistical methods.
>


You'll find them impressive then.

>
> It's these whole population studies that are the junk science. If they
> convince anyone, then those people lack the skill to recognize junk
> science.
>


Funny that. I've just sat on a panel chaired by the Director General of
the UK Research Councils reviewing grant applications. Let me see,
whose assessment of my ability to recognise junk science do I trust
most, the Director General of the Research Councils or Scharf the Larf?
Tough question.

--
Tony

"The best way I know of to win an argument is to start by being in the
right."
- Lord Hailsham
 
SMS wrote:

> James Annan wrote:
>
>> SMS wrote:
>>
>>> As gds points out, the retrospective population studies regarding
>>> helmets are flawed. Studies done pre and post compulsion are especially
>>> flawed, since compulsion introduces a whole new set of variables that
>>> are not, and can not, be accounted for.

>>
>>
>> The fact that these hypothetical variables always mask or even outweigh
>> the "benefit" of a massive increase in helmet wearing is itself proof
>> that helmets have hardly any effect.

>
>
> Two problems with that statement. First, these variables are not
> hypothetical, if you believe the various papers written by the
> anti-compulsion people. Second, those variables are what make
> determining the actual effect of a helmet law so difficult.


....because the "helmet effect" is too small to be seen over them, which
is what I already said.

> If any entity ever decides to do a proper population based study of
> helmet effectiveness, then it should be in a country without compulsion.
> Until such a study is done, if ever, it's important to reject the junk
> science on both sides of the debate, while accepting the raw data on
> helmet effectiveness in accidents.


Crook and Feikh was exactly one such study. I'm sure there have been
legions more.

James
--
James Annan
see web pages for email
http://www.ne.jp/asahi/julesandjames/home/
http://julesandjames.blogspot.com/
 
In article <[email protected]>,
[email protected] writes:

> Any beginner who rolls out with a
> happy-go-lucky I am relatively safe attitude is
> in for it. This is not Holland, dumbasses.
>
> Of course I will be labeled a 'fearmonger' for
> speaking the truth on this.


Not by me. But perhaps I've seen more Stupid Driver Tricks
(and, I must admit -- Stupid Cyclist Tricks) than many
Dutch urbanites.


cheers,
Tom

--
-- Nothing is safe from me.
Above address is just a spam midden.
I'm really at: tkeats [curlicue] vcn [point] bc [point] ca
 
SMS wrote:
> [email protected] wrote:
>
> > The statistics reflect the relatively small amount of
> > cyclists who ride the vast majority of miles. These
> > are riders who already understand the danger
> > of traffic.

>
> This is a very big problem when the various entities do their studies.
> It's a very big variable that is invariably ignored.


To be fair, the deaths of children are often
counted along with those of adults in these per-hour
calculations, which obscures the issue whether
you want to examine what happens with children
or with adults.

Kids' bike fatalities seem to be
on a steep decline, most likely (imo)
because kids' bike riding is on a steep decline.

Robert
 
[email protected] wrote:

> You believe? Wait a minute - aren't you the guy who's forever asking
> for not only data, but all the details on how the data was collected?


No, I'm the guy who points out that your data is
worthless until there is some way we can check
the work. I'm also the guy who says that anybody
who sets out to prove or disprove the danger of
cycling with accident statistics is in for a rough
road and will have to step around a lot of horse
hockey. And I'm also the guy who has sex for reasons
other than procreation while wearing a ballcap.
Oh wait, that's another thread.

> Naturally, I'll expect you to meet your own standards now, and tell us
>he injury rate per mile and per hour for "veteran bike messengers,"
> with full documentation. ;-)


I'm afraid no surveys have been conducted. Come
to think of it, this is a shame, as the population of
bike messengers, let alone long-timers, is pretty
small. Maybe you could help design a survey.
No surveys, but I personally have been well
acquainted enough with the accident and medical
history of perhaps 100 messengers to tell you
a thing or two about how it all shakes out.
For instance, it is obvious that the accident rate
of rookie messengers is astronomical compared
to the veterans. It is also obvious that most of
the veterans ride much longer between serious
wrecks than the 30,000 or so miles of LAB members
in the Moritz survey.


> Compare,
> > for instance, to Moritz' survey of LAB members
> > with an average of 17 years riding experience,
> > which found them injuring themselves about
> > every 30,000 miles (iirc).

>
> Something like that. Skinned knees mostly, remember?


That is incorrect. According to Moritz '96 survey of
LAB members (averaging almost 15 years cycling
experience), almost 10% reported suffering some
*serious* crash in the previous year. These crashes
had an average and mean medical expense of 3000
and 150 dollars respectively. In addition, almost
one third of survey respondents reported
any sort of accident in the previous year.
That's the 'skinned knees' you're talking about,
a much more frequent occurrence. That's
roughly one third of these very experienced
riders wrecking per year--two thirds of the
wrecked end up with skinned knees or
similar superficial injury, one third with
something worse.

http://www.bicyclinglife.com/Library/Moritz2.htm

Interestingly, this corresponds quite closely
with Ken Kifer's much smaller 2001 survey,
most of the respondents to which were touring
cyclists who heard about the survey on Ken's
website. These riders reported a similar level
of experience and average age to those on
Moritz' survey, and reported wrecking (serious
vs. minor) at a very similar rate. Kifer notes
something very important: the average accident
rate among all respondents for what he called
'real injuries' was a seemingly chill once per
23000 miles. But, among those who actually did
the wrecking, their serious wrecks came about
once per 2000 miles.

www.kenkifer.com/bikepages/survey/sept01.htm

> > That's roughly 2-3
> > times the accident rate of the old messengers I know.

>
> Citation?


Read it again. 'Old messengers I know.' I know them
well enough to know their rates of serious injury.

> > Beginning riders wreck/injure themselves at
> > at least 10 times the rate of old messengers (Forester).

>
> You've alluded to this before, but I don't recall the details. What
> exactly did Forester say, and how exactly did he determine it?


This is a compilation from many surveys that
are now outdated, but his contention of a steep
learning curve and vast differences of accident
rate depending on experience level continues to be
corroborated by newer surveys like Moritz, and
the personal experience of high-mileage cyclists
with good memories.

> > To any beginners out there who have themselves
> > convinced that riding in traffic is 'relatively safe,'
> > I will say, It's not safe for you, dumbass.

>
> "Traffic" being defined how, again? Last time I pressed you for a
> definition, you said "If there are any motor vehicles anywhere on the
> road, it's traffic" ... or words to that effect.
>
> IOW, riding on a wide road that's deserted except for you and one other
> car passing the opposite way at 20 mph is riding in "traffic." So
> beginners should be warned that even that is ... dangerous!!! <wring
> hands here>


If passing of the vehicle coincides with an
intersection of any sort, it's plenty dangerous.
Otherwise, not so much.

<wring hands here>

Robert
 
[email protected] wrote:
> According to Moritz '96 survey of
> LAB members (averaging almost 15 years cycling
> experience), almost 10% reported suffering some
> *serious* crash in the previous year. These crashes
> had an average and mean medical expense of 3000
> and 150 dollars respectively.


Mean and average are synonyms so it's not clear how these could differ
by a factor of 20.
Perhaps some other statistical measure was actually reported?
....
> Kifer notes
> something very important: the average accident
> rate among all respondents for what he called
> 'real injuries' was a seemingly chill once per
> 23000 miles. But, among those who actually did
> the wrecking, their serious wrecks came about
> once per 2000 miles.


I hope you realize that this "very important" latter finding is just a
trivial consequence of how the survey is taken. I.e. if you call
people who ride about 2000 miles a year and ask them if they've had a
serious accident in the last year, then of course you'll find that the
subset that did have a crash experienced it within the last 2000 miles.
If instead you asked the same group about any serious accidents in the
last week (or about 40 miles of riding) you'd find that those few who
did have one experienced it in only about the last 40 miles of riding.
Gee what a bunch of daredevil riders they must be! :)

But of course the relevant statistic is the overall rate of an accident
every 23000 miles, not the number you get by just considering the
subset of respondents who were unfortunate enough to have an accident
shortly before you conducted the survey.
....
> > > Beginning riders wreck/injure themselves at
> > > at least 10 times the rate of old messengers (Forester).

> >
> > You've alluded to this before, but I don't recall the details. What
> > exactly did Forester say, and how exactly did he determine it?

>
> This is a compilation from many surveys that
> are now outdated, but his contention of a steep
> learning curve and vast differences of accident
> rate depending on experience level continues to be
> corroborated by newer surveys like Moritz, and
> the personal experience of high-mileage cyclists
> with good memories.


None of that looks like a quote from Forester nor a reference to any of
his writing. I would also be interested in seeing exactly what
Forester supposedly said about the rate of accidents among bike
messengers.
....
> > IOW, riding on a wide road that's deserted except for you and one other
> > car passing the opposite way at 20 mph is riding in "traffic." So
> > beginners should be warned that even that is ... dangerous!!! <wring
> > hands here>

>
> If passing of the vehicle coincides with an
> intersection of any sort, it's plenty dangerous.
> Otherwise, not so much.


Guess I''ve been living very dangerously then and even encouraging my
daughter to do so as well. Sure doesn't seem like it though.
 
peter wrote:

> Mean and average are synonyms so it's not clear how these could differ
> by a factor of 20.
> Perhaps some other statistical measure was actually reported?


Sorry that should read median not mean.


> > This is a compilation from many surveys that
> > are now outdated, but his contention of a steep
> > learning curve and vast differences of accident
> > rate depending on experience level continues to be
> > corroborated by newer surveys like Moritz, and
> > the personal experience of high-mileage cyclists
> > with good memories.

>
> None of that looks like a quote from Forester nor a reference to any of
> his writing. I would also be interested in seeing exactly what
> Forester supposedly said about the rate of accidents among bike
> messengers.


It's in Bicycle Transportation: A Handbook for
Cycling Transportation Engineers, 1994. What
he claimed was that 10 years of riding experience
reduced a rider's injury rate by 80%.

He's never said a word about messengers afaik.

> > > IOW, riding on a wide road that's deserted except for you and one other
> > > car passing the opposite way at 20 mph is riding in "traffic." So
> > > beginners should be warned that even that is ... dangerous!!! <wring
> > > hands here>

> >
> > If passing of the vehicle coincides with an
> > intersection of any sort, it's plenty dangerous.
> > Otherwise, not so much.

>
> Guess I''ve been living very dangerously then and even encouraging my
> daughter to do so as well. Sure doesn't seem like it though.


No, it sure doesn't, most of the time.

Robert
 
[email protected] wrote:

> It's in Bicycle Transportation: A Handbook for
> Cycling Transportation Engineers, 1994. What
> he claimed was that 10 years of riding experience
> reduced a rider's injury rate by 80%.


It's amusing when you see the posts by cyclists in countries with MHLs,
claiming that they've been riding for 20, 30, 40, 50, years, etc., but
gave it up when the MHL was enacted. Then they claim that the number of
cyclists has gone down, solely because of the MHL, yet the
injury/fatality rate has declined only linearly with the alleged decline
in the number of riders. There's at least three problems with their
claims. First, if indeed cycling has declined, it was not due solely to
the MHL. Second, if it's the long time riders that have given up cycling
rather than wear a helmet, then the average experience level of the
remaining cyclists has gone way down, and third, the injury fatality
rate reduction must be attributed to helmets, since it's now less
experienced cyclists, which have more accidents, that are comprising a
larger proportion of the cycling population.

Of course, none of their original claims have been proven, so until a
real study is done, there is no way of knowing any of this.
 
[email protected] wrote:
> peter wrote:
>
> > Mean and average are synonyms so it's not clear how these could differ
> > by a factor of 20.
> > Perhaps some other statistical measure was actually reported?

>
> Sorry that should read median not mean.


Right. And the paper specifically points out that there was one
respondent with a medical expense of $250,000 (i.e. an outlier).
"Clearly such an amount will exert a strong upward bias on the
average." IOW, the median value is much more realistic. The median
"serious" crash cost only $150. That does _not_ sound like the cost of
fixing a broken leg! That's more like what an ER will charge for
putting a band-aid on your finger.

>
> > > This is a compilation from many surveys that
> > > are now outdated, but his contention of a steep
> > > learning curve and vast differences of accident
> > > rate depending on experience level continues to be
> > > corroborated by newer surveys like Moritz, and
> > > the personal experience of high-mileage cyclists
> > > with good memories.

> >
> > None of that looks like a quote from Forester nor a reference to any of
> > his writing. I would also be interested in seeing exactly what
> > Forester supposedly said about the rate of accidents among bike
> > messengers.

>
> It's in Bicycle Transportation: A Handbook for
> Cycling Transportation Engineers, 1994. What
> he claimed was that 10 years of riding experience
> reduced a rider's injury rate by 80%.


Wait, Robert! Not only does that say nothing at all about bike
messengers; that says nothing at all about how the number was
determined!

One of your main arguments is that the "cycling is safe" figures are
worthless unless we know _exactly_ how they were determined! After
disparaging the "safe" data, it's VERY hypocritical to put up such a
vague paragraph - let alone to keep quoting the data as frequently as
you do!

ISTM your standard is really this: If it makes cycling sound
dangerous, no proof is needed; if it makes cycling sound safe, no proof
is sufficient.

- Frank Krygowski
 
SMS wrote:
>
> It's amusing when you see the posts by cyclists in countries with MHLs,
> claiming that they've been riding for 20, 30, 40, 50, years, etc., but
> gave it up when the MHL was enacted. Then they claim that the number of
> cyclists has gone down, solely because of the MHL, yet the
> injury/fatality rate has declined only linearly with the alleged decline
> in the number of riders. There's at least three problems with their
> claims. First, if indeed cycling has declined, it was not due solely to
> the MHL. Second, if it's the long time riders that have given up cycling
> rather than wear a helmet, then the average experience level of the
> remaining cyclists has gone way down, and third, the injury fatality
> rate reduction must be attributed to helmets, since it's now less
> experienced cyclists, which have more accidents, that are comprising a
> larger proportion of the cycling population.
>
> Of course, none of their original claims have been proven, so until a
> real study is done, there is no way of knowing any of this.


Those are some interesting observations. Particularly about what the
experience distribution of cyclists abandoning cycling in response to
MHL's might be. I have no idea what that distribution might be but
certainly if it varies from the overall experience distribution of pre
MHL cyclists then before vs. after injury rate copmparisons are
problmeatic even if non reported accidents are somehow addressed.

Interestingly if you look at many groups of very experienced cyclists,
by a variety of measures, you are very l ikely to find that many (of
us) now use helmets but certainly if you are over the age of , say, 50
it is very likely that lots, even most, of that cycling experience was
gained helmetless.

Just like the idea of risk compnesation, which is a valid concept but
very hard to measure for this question, so is the idea of what
behaviorial differences might exist between adaptors and non adaptors
of helmets. And since these variables are not well addressed in the
extant studies we simply don't know. And given boththe relatively low
accident rates and the very low rate of "who cares" it is likely that
we will never have a really good handle on this unless some motivated
individual or organization actually goes out and sponsors the conduct
of a proper study.
 
[email protected] wrote:

> > Sorry that should read median not mean.

>
> Right. And the paper specifically points out that there was one
> respondent with a medical expense of $250,000 (i.e. an outlier).
> "Clearly such an amount will exert a strong upward bias on the
> average." IOW, the median value is much more realistic.


Yes! Any nasty injuries or particularly bad
experiences of individuals must be expunged
from our accident survey, to make 'more
realistic' results! Clearly these 'outliers' will
'exert a strong upward bias on the average'
so we shall ignore them lest they jack up
our surveys! Careful readers of the Kifer
survey, for instance, may notice that he has
chosen not to report the injuries of one
respondent who claimed 4 injuries in the
previous year. Of course, the experiences of
any who ride 50 years with no injury will
be gleefully included without an eye-blink.
Stats are fun aren't they?

> The median
> "serious" crash cost only $150. That does _not_ sound like the cost of
> fixing a broken leg! That's more like what an ER will charge for
> putting a band-aid on your finger.


If the respondents are anything like the riders
I know, several of these serious wrecks resulted
in rather substantial injury for which the rider
(stupidly perhaps) sought absolutely no medical
treatment. If he had, it might have cost thousands--
x-rays, physical therapy, etc.

> > It's in Bicycle Transportation: A Handbook for
> > Cycling Transportation Engineers, 1994. What
> > he claimed was that 10 years of riding experience
> > reduced a rider's injury rate by 80%.

>
> Wait, Robert! Not only does that say nothing at all about bike
> messengers; that says nothing at all about how the number was
> determined!


It says a great deal about bike messengers.
He's saying experience reduces accident rates.
Veteran messengers have more experience in
traffic than any other type of cyclist. A commuter
would have to ride about two and a half hours
each way, every day, year round, year after year
to keep up with a full-time messenger
in terms of hours in the saddle. And I'm saying
that yes, Forester's claim does seem to be
corroborated in real life, and it does seem
generally applicable to the whole spectrum of
cyclists, from the kids right up to the old bike
messengers, even if Forester did not explicitly
address bike messengers at the time. The
exact nature of the curve is in question, but the
fact that it's quite a hill is obvious at this point.
I doubt Forester himself would acknowledge
that messengers may actually be the safest
riders on the road because it would call into
question his entire philosophy and the Effective
Cycling industry that has grown up around it.

Forester's claim is a product of compilation of
statistics from several surveys, Schupack, et al,
Kaplan '76, Watkins' survey of CTC riders.
Tricky business to be sure! But the statistical
importance of experience continues to be
supported in newer surveys like Kifer and Moritz.

> One of your main arguments is that the "cycling is safe" figures are
> worthless unless we know _exactly_ how they were determined! After
> disparaging the "safe" data, it's VERY hypocritical to put up such a
> vague paragraph - let alone to keep quoting the data as frequently as
> you do!
>
> ISTM your standard is really this: If it makes cycling sound
> dangerous, no proof is needed; if it makes cycling sound safe, no proof
> is sufficient.


I don't get it. What about these surveys above says
'cycling is safe' or 'cycling is dangerous?' [1] It's clearly a
more complex picture than you make it out to be.

Why do you think Forester's claim that 10
years of cycling reduces the accident rate by
80% makes cycling sound dangerous? That's not
exactly what I would take away from that number.
To me that says we have the power to control
our own fate while on the bike, that it's not some
random crapshoot of falling and getting hit by
cars. If it were, accident rates would be normalized
across experience levels, and clearly they are not.
This is good news! Or bad news, depending on
your level of experience.

Robert

[1] Other than Kifer's finding that 'injury is 33 times
more likely to occur from riding a bike as opposed
to driving a car for the same distance.' I admit
that does make cycling sound dangerous.