I emailed Jan about this discussion and he caught up through the archives.
He provided this reply and asked me to forward it back to the list:

---------- Forwarded message ----------
Date: Thu, 17 Jul 2008 10:44:28 -0700
From: Jan Heine <heine94@...>
To: Alex Wetmore <alex@...>
Cc: Mark Vande Kamp <mevk@...>
Subject: Re: [KOG] Re: KOG: RAGBRAI? (fwd)

We - Bicycle Quarterly, framebuilder Jeff Lyon, True Temper and Henry
James (lugs) - spent a lot of money and a lot of effort to make sure
this test was indeed double-blind. We had observed planing in the
past, and correlated it to frame tubing, but there always were
extraneous factors that we could not control. (The best-planing bike
was red! Red bikes always are faster, aren't they?) By creating three
outwardly identical bikes, with only the top and down tube wall
thicknesses as a difference, and the weights compensated between the
frames, we really could eliminate all other factors. And by making it
double-blind, we eliminated our biases as well.

During our tests, it was very clear that the "planing" bikes were
faster for Mark and me. When one of us was on the superlight bike, it
was easy to drop the other on the "standard" bike. And when both were
on superlight bikes, we were well-matched.

Before somebody says that this is unscientific, two well-matched
riders provide a more objective measure of bike performance than
measuring speed, heart rates or power output, all of which are
affected by increasing fatigue, wind and other factors. Two
well-matched riders fatigue at the same rate, and they ride
side-by-side, so the wind is the same, too. If one draws ahead, and
then you switch bikes, and the roles are reversed, and you then
shuffle the cards (by switching the stem caps around), and the same
result is repeated, you know that one bike is faster _for these
riders_ than the other.

Since we determined the stiffness of the frame based on its
performance - we did not feel frame flex, but we deducted it based on
the frame's performance - it really was a double-blind test of
bicycle performance and planing. Perhaps we aren't sensitive enough,
or we just don't care, but neither of us noted any difference in BB
flex or elsewhere on these bikes. The handled the same, we did not
notice any rubbing of chains on derailleurs or other giveaways. The
only difference was that they performed differently.

We have been grappling with trying to come up with an instrumented
measure of planing, and we have thought about heart rate, power
output, speed and other parameters that could be measured.

However, our current hypothesis is that planing allows you to work
harder with less lactic acid buildup. Your heart rate is higher, your
power output is higher, you are more fatigued at the end of the ride,
but you have been able to maintain a higher power output than you
would have if the lactic acid buildup was the limiting factor.
Unfortunately, lactic acid buildup is hard to measure on the bike. We
are open to ideas here.

Planing is not magic, like a small motor built into your rear hub.
It's not a free lunch, it's just an optimization of the interplay
between human body and machine. When the original finite element
analysis article was peer-reviewed, one reviewer pointed out that the
model did not account for the rider to get in sync with the bike. If
you do not get in sync with the bike, then the benefits of planing
are too small to notice. This is like dribbling a basketball by
hitting it randomly, rather than at just the right time, or trying to
swing on a swing set by "pumping" randomly. If you get in phase with
the ball/swings, the same effort sees very different results.

Another issue is that now we know which bikes work best for Mark and
I. We do not really know what works best for Alex or the millions of
other riders out there. All we have shown is that planing exists,
that it is related to frame flex, and that it makes sense for riders
to try bikes with different stiffnesses to see what they like best.
We never will achieve absolute perfection, but when given a choice
between a "sturdy OS" frame like some I have tested and the
"standard" frame that planed less in our test, I know I would be
perfectly happy with the "standard" frame. In fact, I have been
riding a frame made from similar tubing for years...

Jan Heine
Editor
Bicycle Quarterly
140 Lakeside Ave #C
Seattle WA 98122
www.bikequarterly.com

>The discussion has started over on the Kogswell list. Are you
>subscribed? If not I can forward messages to you.
>
>alex
>
>---------- Forwarded message ----------
>Date: Thu, 17 Jul 2008 09:46:10 -0700
>From: james black <chocotaco@...>
>Reply-To: "KOG@yahoogroups.com" <KOG@yahoogroups.com>
>To: "KOG@yahoogroups.com" <KOG@yahoogroups.com>
>Subject: Re: [KOG] Re: KOG: RAGBRAI?
>
>On Thu, Jul 17, 2008 at 9:18 AM, alex wetmore <alex@...> wrote:
>>My results were random (I only identified the correct bike on the first
>>day). Jan and Mark were able to identify the correct bike each day. I
>>had knee issues during tests #3 and #4 which prevented me from riding
>>hard and I think that is why I had trouble identifying the right bike
>>on those days. I also don't ride as hard as them and this may have
>>made it harder for me to detect.
>
>I have been reading BQ for a couple of years now, and I have been very
>skeptical of planing. This blind test left me wanting more. It seems
>like there must be a way to do something more scientific that makes
>use of the three identical bikes.
>
>That is, my take on the article was that it demonstrates that in fact,
>two out of three riders could identify the more flexible bike blind,
>therefore planing is real. The fact that they could identify the
>flexier bike does not mean that flexy bikes inherently go faster, and
>the article doesn't really address that - only in subjective accounts
>by riders of bikes feeling fast or sluggish.
>
>I thought it was great work that BQ demonstrated in its tire rolldown
>tests that there can be tires that feel fast but are in fact slow, and
>vice-versa; that subjective feelings can be liars, and by extension,
>when someone tells you that their tires are fast, they may have been
>fooled by their senses. Some may say that a tire that feels fast is
>inherently valuable, but I say that it is better to have a tire that
>feels slow and goes fast than one that feels fast and goes even a
>little less fast.
>
>So how do we know that the "planing" bikes don't just feel fast, or
>have a responsiveness that is associated with being fast? Could it be
>that the flexible bike seems faster because it somehow encourages
>harder riding? Some might say that this would be an inherently
>valuable trait in a bike, but I would not.
>
>This test proved that flexibility could be sensed by the rider in a
>blind comparison. Could there be another test that establishes that
>the flexible bike is faster, based on other than subjective account?
>Say, have the rider wear a heart rate monitor and do a typical ride (x
>laps around a velodrome at a constant speed or something) and evaluate
>the effort required to accomplish the same work?
>
>At one point I wondered about those power tester hubs or cranks or
>whatever they are, but someone thought that these would not account
>for planing because the extra inputs provided by planing would go back
>through the power measurer device.
>
>One more thing: the recent BQ test referenced an finite element
>analysis by some guy that was in BQ a few issues back, and which
>purported to refute the claim that flexible frames are slower than
>stiff frames. The analysis asserted that as much as half a percent of
>rider energy might get taken up in flexing of the frame, but that most
>of this would get returned into forward motion after the frame
>returned from its deflection. The only loss would be heat loss in
>friction within the frame.
>
>Supposing for a moment that this entire .5% of force was like a bonus,
>entirely returned during that hole in between power strokes in the
>pedal cycle, then this analysis would suggest that as much as .5%
>improvement in power could be generated by planing. That's not much.
>Would it even be detectable? I can't discern the difference between my
>speed, effort, or exhaustion with that kind of accuracy. The effect of
>.5% would be vastly less than the effect of a change in tires or a
>slight change in aerodynamics.
>
>There are a lot of reasons to be skeptical about the concept of
>planing, and I think the advocates for it have an uphill battle to
>explain how it works, and show how it could be big enough to be
>significant. I don't want to discount the opinions of Jan, Mark and
>Alex, three of the smartest bike guys around; but I haven't seen a
>compelling case for planing yet.
>
>James Black
>
>------------------------------------
>
>Yahoo! Groups Links
>
>
>