Re: I crash into religion

[Chris Slade]

Tony Raven wrote:

> David Martin wrote:
>> jtaylor wrote:
>>> Why do you think this leads to a better decision?
>>
>> Multiple presumptions in that question..
>>
>> For those who believe, no explaination is necessary. For those who do
>> not believe, no explaination is possible.
>>
> How do you classify those of us who used to believe until we reviewed
> the evidence and currently do not believe but remain open to new evidence?

Heretics and apostates who have strayed from the one true (singletrack?)
path of righteousness!

Or as Peter put it, AHZs.
--
Chris

I'm going to live forever, or die trying!
-- Spider Robinson

 

[David Damerell]

Quoting p.k. <[email protected]>:
>David Damerell wrote:
>>No, as I've already explained, and as you're carefully dodging away
>>from. If we have _zero_ information about the risk distribution the
>>population statistics are absolutely the best available guide to an
>>individual's risk and the rational thing for an individual to do is
>>to act as if the population statistics apply to them.
>I don't think the OECD agree with you:

If you do not comprehend the statement in my article you are not competent
to comment on it. Sorry.
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[David Damerell]

Quoting Espressopithecus (Java Man) <[email protected]>:
>[email protected] says...
>>No, as I've already explained, and as you're carefully dodging away from.
>>If we have _zero_ information about the risk distribution the population
>>statistics are absolutely the best available guide to an individual's risk
>>and the rational thing for an individual to do is to act as if the
>>population statistics apply to them.
>Pedantic and impractical. If we all lived on the basis of being unable
>to make decisions without adequate data, we'd never leave the house.

Which is a neat distraction, since of course - lacking any other data - we
have the population statistics to let us make decisions.

>Come to think of it, we'd never buy the house in the first place since
>information cannot predict the future value of homes,

Actually there is a great deal of information available about the likely
future value of houses and furthermore we might require a house for some
purpose other than its future value.
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[David Damerell]

Quoting Espressopithecus (Java Man) <[email protected]>:
>As I've said in other posts, feel free to make your own decisions about
>which gross population level statistics to apply to your particular
>situation, and which not. I'm not trying to make a case for everyone
>disregarding all statistics -- I'm trying to make a case for making good
>decisions in the absence of statistics that bear directly on the
>question at hand.

Now, how might we know if that decision is good?

If the population statistics show zero net effect for helmets, but your
deduction is that nevertheless you are in a group that gets a positive
effect, there are three questions to answer.

1) What's the group that gets a negative effect?

2) Given the idea we have of the effects that cause helmets to have a net
zero effect - rotational versus direct impacts, etc - do those seem like
the sort of thing that one type of riding might be particularly prone to?

3) Given that everyone who makes this deduction imagines they are in a
group that gets a positive effect, they can't all be right. What's special
about you?
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[David Damerell]

Quoting Espressopithecus (Java Man) <[email protected]>:
>[email protected] says...
>>Quoting Espressopithecus (Java Man) <[email protected]>:
>>>I agree that the data isn't there. And in the absence of data, I'll
>>>trust my judgement over a crude population mean.
>>Even though the world is full of people trusting their judgement and,
>>given that the population mean effect of helmets is about zero, if there
>>exist people whose judgement is good, there has to be a corresponding set
>>of people whose judgement is bad?
>Yes, and those who think a crude population mean is a good substitute
>for mature judgement based on experience are among them.

When you resort to cheap shots it only serves to highlight the fact that
you are not answering the question.

Please do so.
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[David Damerell]

Quoting Espressopithecus (Java Man) <[email protected]>:
>[email protected] says...
>>Quoting <[email protected]>:
>>>If your biking is mostly low-speed mountain biking, I think it really
>>>is another kettle of fish.
>>Well, quite. I don't want to talk about off-roaders, really; I don't think
>>many of the large-scale studies pertain to off-road riding, I don't think
>>it can tell us much about on-road riding, and while I'd oppose compulsion
>>for off-roaders I think the benefits of foam hats there may well be
>>positive.
>How does this square with the two statements below, which you recently
>posted:
>"No, actually; I regard it [that there are cycling activities that
>present more risk of the kinds of head injuries against which helmets
>provide protection] as best not proven.

Well, I think "may well be positive" and "not proven" are not
contradictory.

>"If we have _zero_ information about the risk distribution the
>population statistics are absolutely the best available guide to an
>individual's risk and the rational thing for an individual to do is to
>act as if the population statistics apply to them."

This is a simple statement of fact. "If we have zero information then X"
does not imply "We have zero information".
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[David Damerell]

Quoting Espressopithecus (Java Man) <[email protected]>:
>[email protected] says...
>>Feel free to counter this. Come up with some hypothetical population and
>>risk where there is a better individual strategy than assuming the
>>population mean.
>I think you've already done it for me:
>"while I'd oppose compulsion for off-roaders I think the benefits of
>foam hats there may well be positive"

You're evading the point again. First of all, "may well be" does not mean
"are"; secondly, that hypothetical excercise is where there is no
information about risk distribution. When I talk about off-roaders I am
speculating about risk distribution. What do these have to do with each
other?
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[David Damerell]

Quoting p.k. <[email protected]>:
>David Damerell wrote:
>>For example, imagine that you have a lottery ticket. Before the
>>lottery is drawn, the "risk" of winning the lottery is evenly
>>distributed across all ticket holders. After it is drawn, the risk is
>>very unevenly distributed. Nevertheless, before you know the result,
>>your behaviour should not change after the draw.
>again you are assuming that there is a random finger of fortune (cf uk
>lottery publicity) that picks out some random ticket owner.

When you have no information about risk distribution that is exactly how
it works.

Please don't argue with statements of the form "If you have no data about
risk distribution then X" by arguing "We have data about risk
distribution".
--
David Damerell <[email protected]> flcl?
Today is Oneiros, June.

 

[Sorni]

David Damerell wrote:
> Quoting p.k. <[email protected]>:
>> David Damerell wrote:
>>> No, as I've already explained, and as you're carefully dodging away
>>> from. If we have _zero_ information about the risk distribution the
>>> population statistics are absolutely the best available guide to an
>>> individual's risk and the rational thing for an individual to do is
>>> to act as if the population statistics apply to them.
>> I don't think the OECD agree with you:
>
> If you do not comprehend the statement in my article you are not
> competent to comment on it. Sorry.

Classic Damnitall.

 

[Sorni]

David Damerell wrote:
> Quoting Espressopithecus (Java Man) <[email protected]>:
>> [email protected] says...
>>> Quoting Espressopithecus (Java Man) <[email protected]>:
>>>> I agree that the data isn't there. And in the absence of data,
>>>> I'll trust my judgement over a crude population mean.

>>> Even though the world is full of people trusting their judgement
>>> and, given that the population mean effect of helmets is about
>>> zero, if there exist people whose judgement is good, there has to
>>> be a corresponding set of people whose judgement is bad?

>> Yes, and those who think a crude population mean is a good substitute
>> for mature judgement based on experience are among them.

> When you resort to cheap shots it only serves to highlight the fact
> that you are not answering the question.

That was a cheap shot?!? (I've moved it down a line from the body of
blather to which you left it blended so the reader can more easily see it.
That's called coherent and considerate posting, Damnitall.)

> Please do so.

He did. (Read it again; question and answer are still right there.)

B

 

[Espressopithecus (Java Man)]

In article <iM*[email protected]>,
[email protected] says...
> Quoting Espressopithecus (Java Man) <[email protected]>:
> >As I've said in other posts, feel free to make your own decisions about
> >which gross population level statistics to apply to your particular
> >situation, and which not. I'm not trying to make a case for everyone
> >disregarding all statistics -- I'm trying to make a case for making good
> >decisions in the absence of statistics that bear directly on the
> >question at hand.
>
> Now, how might we know if that decision is good?

Without additional data, we won't, as with you and wine drinking, and as
with those athletes whose so-called "risky" BMIs were, by the best
population data available, put them at higher risk of cardiovascular
disease. Of course, that was before further studies showed that fit,
muscular men with high BMIs were NOT at increased risk despite the
overall population findings. But you would have them . . . do what? .
.. . stop exercising until their BMIs were in line with the best
available whole population evidence?
>
> If the population statistics show zero net effect for helmets, but your
> deduction is that nevertheless you are in a group that gets a positive
> effect, there are three questions to answer.
>
> 1) What's the group that gets a negative effect?

You've already addressed that.

---------------
"Well, quite. I don't want to talk about off-roaders, really; I don't
think many of the large-scale studies pertain to off-road riding, I
don't think it can tell us much about on-road riding, and while I'd
oppose compulsion for off-roaders I think the benefits of foam hats
there may well be positive."
---------------

If off-roaders aren't included in the population stats, why does there
have to be a group that "gets a negative effect"?
>
> 2) Given the idea we have of the effects that cause helmets to have a net
> zero effect - rotational versus direct impacts, etc - do those seem like
> the sort of thing that one type of riding might be particularly prone to?

I don't know, but I agree with your assessment above -- "for off-roaders
I think the benefits of foam hats there may well be positive".

>
> 3) Given that everyone who makes this deduction imagines they are in a
> group that gets a positive effect, they can't all be right. What's special
> about you?
>
What's special about me? Nothing, but I do try to keep up.

Rick

 

[Espressopithecus (Java Man)]

In article <Mnj*[email protected]>,
[email protected] says...
> Quoting Espressopithecus (Java Man) <[email protected]>:
> >[email protected] says...
> >>Feel free to counter this. Come up with some hypothetical population and
> >>risk where there is a better individual strategy than assuming the
> >>population mean.
> >I think you've already done it for me:
> >"while I'd oppose compulsion for off-roaders I think the benefits of
> >foam hats there may well be positive"
>
> You're evading the point again. First of all, "may well be" does not mean
> "are"; secondly, that hypothetical excercise is where there is no
> information about risk distribution. When I talk about off-roaders I am
> speculating about risk distribution. What do these have to do with each
> other?
>
The majority of my riding is off-road. You already agreed that the
population stats probably don't apply:
----------------------------------
[email protected]>:
"If your biking is mostly low-speed mountain biking, I think it really
is another kettle of fish."

You:
"Well, quite. I don't want to talk about off-roaders, really; I don't
think many of the large-scale studies pertain to off-road riding, I
don't think it can tell us much about on-road riding."
-----------------------------------


And you speculated:
------------------------------------
" . . while I'd oppose compulsion
for off-roaders I think the benefits of foam hats there may well be
positive."
------------------------------------

I agree that the benefits of helmets off-road may well be positive, and
in the absence of data pertaining to off-road riding, I choose to ride
helmeted.

Rick