On Sun, 18 Feb 2007 21:54:17 -0500, John Forrest Tomlinson
<[email hidden]> wrote:
[snip]
Quoted message said:So how is it that I can set up a new frame, with the same seatpost and
seat from my old bike (and cranks, and BB and pedals), in the same
clothes that I rode on my old bike, ride it once and feel my knee is
tight at the end of the ride, then look really carefully at it and see
either the seatpost slipped or I made a tiny error in measurement.
Am I imagining it?
Well I guess I must be, huh.
[snip]
Dear John,
Yes, you probably are. No, you probably won't believe it.
A classic sign of princess-in-the-pea syndrome is the belief that a
different pair of pants or slightly thicker socks could cause knee
problems.
Doctors are used to patients who insist that tinier and tinier changes
have amazingly large effects.
It's hard to remember that these hypochondriac patients are sincere.
They're so busy trying to identify tiny changes in how they feel that
they both magnify whatever they happen to feel and equate it with
whatever recent change they happen to have made.
The basis of most snake oil remedies is that the patients who happen
to improve quite naturallly give all the credit to whatever elixir or
therapy they were using at the time.
Poor Eddy Mercx can be seen in videos, stopping to fiddle with his
seat height during Paris-Roubaix. Given the pain of his back injuries,
it's understandable that he constantly adjusted his seat height in
hopes that something would help.
But it's just an example of what W. Edwards Deming illustrated with
his funnel-and-marble demonstration.
If you set up a funnel on a stand and drop one marble through it, you
can see where the marble lands near a target after it rattles through
the wide nozzle of the funnel.
If you move the funnel-stand after every drop to try to make the next
marble hit the target, you get a much worse result than you would if
you averaged a dozen marbles.
You simply can't do any better than the average determined by how wide
the funnel is. Each marble hits randomly somewhere inside a ring.
Chasing each random hit only widens the average concentration. Any
target shooter is painfully aware of this.
Unfortunately, that's what we often do because of practical limits. We
want to ride faster or do something about knee pain, so we try a
different [seat/crank/handlebar/frame/gear/shoe/pedal/whatever].
Normal random variation means that in the next two weeks we're likely
to have a [sore-back/fast-time/tight-knee/happy-ride].
We mistakenly attribute the random result that would have happened
anyway to whatever we happened to be adjusting at the time, and we
refuse to test long enough to eliminate random variation.
It's much quicker to do things this way than to spend months
collecting data, and it's very satisfying to know that we are [very
sensitive/slightly faster/avoiding injury/better equipped].
The process works best where things matter least.
No one is immune to this sort of thing. No matter what I think, I
can't imagine using anything except 175 mm cranks--anything shorter
would surely be inefficient, while anything longer would undoubtedly
flex my joints painfully!
(Luckily, I never looked at the cranks on the bikes that I rode for my
first 50,000 miles--they probably had 172.5 mm or even 170 mm cranks
and must have been terrible.)
To complicate things, chasing random variation and self-exaggerated
illnesses can turn out to be right--there's a faint possibility that
your knee is so delicate that a crank a tenth of an inch longer will
ruin it. But normal knees happily accept much larger variations.
People who fit things are cursed by this problem. Any small change
that they make may well get the blame or credit for something
completely unrelated.
An awful lot of bike-fit theories suggest Langmuir's laws of bad or
pathological science:
1. The maximum effect that is observed is produced by a causative
agent of barely detectable intensity, and the magnitude of the effect
is substantially independent of the intensity of the cause.
2. The effect is of a magnitude that remains close to the limit of
detectability; or, many measurements are necessary because of the very
low statistical significance of the results.
3. Claims of great accuracy.
4. Fantastic theories contrary to experience.
5. Criticisms are met by ad hoc excuses thought up on the spur of the
moment.
6. Ratio of supporters to critics rises up to somewhere near 50% and
then falls gradually to oblivion
http://www.cs.princeton.edu/~ken/Langmuir/langB.htm#Characteristic%20Symptoms
Most people are quite comfortable with left and right shoes of the
same size, even though they have left and right feet that are
measurably different in length and width on a shoe-store scale.
I hope that's enough words for you.
Cheers,
Carl Fogel