I couldn't find that definition of the word anywhere but let's assume it is correct. Perhaps you could tell us the constituent parts of this aerodynamic gain and how large each of them are adding up to 100% of the whole.

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

I guess you missed post #305 where I showed you that using a VERY conservative assumption (i.e. that the wind conditions had similar effects in both years...which they didn't, 2008 was worse) that the power only accounted for ~1/2 of the apparent "gain" (the total being ~1 minute for conditions and 1:45 in time, for a total "gain" of 2:45). IMO, the wind was actually worse than that so it would be a "gain" attributable to the aerodynamics of more than 1:25. Umm...let's see, 85s divided by 37.1 km comes out to...2.3 s/km...which would "equate" to ~.020-.025 m^2 of improvement in CdA.

BTW, I avoid using tools that don't work...

---

http://bikeblather.blogspot.com/

I think the difference is that I used the CdA calculated by Alex Simmons in post 290 where in 305 you were still using the original CdA of your test. I thought there was agreement that was the better number for you time trial. Either way, we are same order of magnitude (these are nuances) and your race would not have been near as good if you had not had those power improvements also.

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

What are you riding this year? Also, does the bet rely on you producing identical wattage?

The suggestion on this thread has been that, all else being equal, the difference in aerodynamics between the P3C and P2K frames results in 2 s/km time savings. This is only supported by Tom's data if we accept that all of the .023m^2 is due to frame aerodynamics alone and not rider positional changes or power meter inaccuracies. We have to be able to attribute all of the delta CdA to the frame alone for it to be true. This is where I think Tom's study design is weak. Without blinding we don't know whether he made systematic changes to his position to make it more aerodynamic on the P3C, therefore we can't attribute all of the delta CdA to frame aerodynamics alone, which invalidates the 2s/km estimates.

As roady has introduced to the discussion, the power meter may have read differently in early and late runs and, since Tom didn't alternate runs, this would introduce major systematic bias in favour of the P3C.

Thanks for this! It suggests what I suspected was true: changes in position on an identical set-up can result in changes in aerodynamics that are of the magnitude of those Tom reported. As someone with no experience of wind tunnel testing I couldn't come up with any data to back it up, so thanks for the help. I tried to ask Dr Chung whether changes in position on an identical set-up could have this magnitude of effect, but he decided not to answer.

I'm not suggesting that Tom manipulated the results by deliberately changing position; I just think there is a whole lot of potential in a study design like this for the subject/experimenter to find what they're expecting to find, whether it's objectively accurate or not. Here are a couple of Wikipedia articles that briefly summarise some of the issues that a study like this fails to effectively deal with:

http://en.wikipedia.org/...er-expectancy_effect

http://en.wikipedia.org/...ct-expectancy_effect

In Tom's study, since the subject and observer were the same guy, the potential for his results to be affected by the above phenomena is enhanced.

My background is in psychology and medicine, fields where the above effects are acknowledged and studies are designed to effectively deal with them. I just find it interesting when I see a study in another field that appears to be open to these biases and yet they are not dealt with effectively by the study design, and experts in the field refuse to acknowledge them.

Glad I'm entertaining someone as well as aggravating Drs Chung and Coggan!

1. You know this phrase "all else being equal?" I don't think it means what you think it means.
2. Once again, you're reverting to the claim that any error invalidates all difference.

BTW, Tom did indeed do alternate runs, as he usually does. It's through the alternate A-B-A runs that we know how much repeatability he has in holding his position: he's done tests where he's sandwiched a water bottle run with two bare frame runs: the bare frame runs differed by .001 m^2. In this case, he said early on in this thread that the wind started to pick up which spoiled the last P3C run. This method is sensitive enough to show when the wind changes.

As for Bob C.'s positional change, positional change could certainly account for a change of this magnitude. However, Bob C did not hit on his turtling spontaneously, he noticed the difference immediately, and he said he could not hold that position for long. You would have us believe that Tom spontaneously found a new position, that he did not notice it, that he held it consistently throughout the P3C run and then lost it for the subsequent P2K run, plus that it was so effective that it reduced (rather than increased) his overall CdA by 10%. How appropriate that your background is psychology, because this argument works is in your mind.

You know, here's another approach: you could calculate the "still air equivalent CdA" for the 2007 and 2008 races.

You know the net elevation gain (around 50 m?).

Let

Jtot = total joules expended = sum(watts)*1.26
JPE = total joules for PE = 84 kg * 9.81 * 50 meters
JKE = total joules for KE = 84 kg * 0.5 * (v[end]^2 - v[start]^2) (v[start]=0, right?)
Jrr = total joules for rr = Crr * 84 kg * 9.81 * 37100 meters (37100 = sum(v)*1.26, right?)
daero = rho * sum(v^3)*1.26 / 2

Then
cda.0 = still air equivalent CdA = (Jtot - Jrr - JKE - JPE)/daero

This is the CdA that would have gotten you around that course in that time with that power had there been no wind. Compare cda.0 to your measured CdA's. You could also construct the still air equivalent VE profiles. They both ought to end up at 50 m, but (I suspect) will have slightly different shapes.

Isn't there also the placebo effect: http://en.wikipedia.org/wiki/Placebo

Bottom line is that our subconscious brain controls much of what we do. I certainly believe Tom was faster on the P3C and my guess is that the difference was is part due to frame aerodynamics. But a subconscious position change cannot be ruled out. He may have felt he was in the exact same position, but some subtle change in setup -- 1 mm difference in armrest or seat height or something -- may have contributed to that feeling. Second, I don't think our bodies are sensitive enough to detect subtle changes in position. Could be that something about being on the P3C induced Tom to make a consistent slight position change.

Now, having said all that, I'm going to ebay to look for P3C's ;) (Actually, don't have the cash (and cash flow is uncertain until Mo St. TT results come in) but am thinking of getting back my old P2, which I loaned out.)
.

---

________
It doesn't really matter what Phil is saying, the music of his voice is the appropriate soundtrack for a bicycle race. HTupolev

What, starting to rethink the bet? ;-)

As you might expect, I'll be riding a P3C, and no, the bet isn't dependent upon identical power (or identical anything else). However, the highest power I've ever sustained for a full 40 km is 297 W, and in 2004 the weather was about as good as it gets on that course. I'm therefore not counting on more power and/or better weather to help me achieve my goal.

In wind tunnel tests conducted at TAMU, the P2 had significantly more drag than my Hooker (complete with their proprietary "aero-or-die" handlebars, which provided a large advantage), which in turn had more drag than my wife's P3C.

I disagree with the bolded statement above. For example, most people can feel a 2.5 mm difference in crank arm length, especially when compared back-to-back (as Tom did his testing).

Mostly I think you're making a fool of yourself. It's all well and good to be skeptical, but taken to the extreme (as you and others have done) it just becomes close-mindedness. After all, it's not as if anyone has ever presented any quantitative data that directly conflicts with Tom's or my observations. IOW, everything - wind tunnel tests, formal field tests using a powermeter, back-calculation from TT performances - point to there being a significant difference in drag between the P2k and the P3C. Yet, you and others still refuse to believe it, just as some deny the existence of evolution or global warming.

I already told you that I wasn't going to break it down for you. After all, you've repeatedly demonstrated that you don't understand physics well enough to follow such an analysis, so why should I waste my time?

careful now - he's gotten really sensitive lately :-)

You took the words right out of my mouth. :-)

That's an interesting idea...but one that's going to have to wait until I have a bit more time to play around with it...

---

http://bikeblather.blogspot.com/

I would disagree. I sold a Velotron with the adjustable frame to a center to do testing on athletes. They also got our adjustable cranks so they could adjust the bike to be just like their regular bike. I visited them after they had it for 6 months and found out talking with them the person doing the testing didn't understand how to tell how long the cranks were, they were using the wrong scribe when setting crank length so everyone was being tested on cranks 10 mm longer than they usually rode. Not one person, apparently, ever noticed anything strange despite the cranks being 10 mm long.

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

they must specialize in serving perceptual idiots ...

"You would have us believe that Tom spontaneously found a new position, that he did not notice it, that he held it consistently throughout the P3C run and then lost it for the subsequent P2K run, plus that it was so effective that it reduced (rather than increased) his overall CdA by 10%. How appropriate that your background is psychology, because this argument works is in your mind."




Why yes, your honor, I DO also believe that pigs fly! Why do you ask?

;-)

Perhaps, or perhaps most of us are not as sensitive to some of this stuff as we imagine. We frequently have different size cranks on our demo bike at expos. It is rare that anyone comments on the crank length, even though most find them hard. In fact, about 50% who get on don't even notice that both legs are down, when I ask them when they first get on, as I am strapping in their second foot, "is there anything strange about the bike that you have noticed yet?"

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

"Largely" is a physics term? I clearly am weaker in that discipline than I thought. :-)

I suspect the real reason you won't break it down is because you can't, and still save face. "80 watts of power savings "largely" due to the frame." you said it, not me.

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

Hmm... While I wouldn't completely discount the possibility of PM error, given the hub you used for the testing I'd bet that it's unlikely. So, unless you were just trying harder on the P3C, that really leaves one likely alternative: you're lying and you made the whole thing up--probably so more people would buy Cervelos and not have any money left for PowerCranks! My money is on this one.

After all, look at what pros do. It's not like riders with P3C's are at a competitive advantage or anything. I mean, it's not like 40% (at least) of the top 10 are riding P3C's or anything--despite the fact that a couple of those riders aren't sponsored by Cervelo. Hmmm.......

Ok, but a P2 that I own only costs me the time to build it. I disassembled and loaned it out because I got frustrated with horizontal dropouts that would slip if I didn't tighten enough (leading to wheel rub on occasion), PIA rear wheel alignment (no set screws), and crappy shifting (problems with internal cable routing). I love riding the replacement, an Aluminum Principia that I bought for $150. It's stiff and solid.

While I'm hardly in the position to ask favor of you, given my contributions above :) I'd appreciate if you (or Tom A.) would be willing to give a SWAG on 40K TT time differential between following two setups:

P2 650c 56 or 57 cm size, with carbonaero fork

Principia Tri26 (fat 1.593" downtube, eggshaped but with 1 cm wide square trailing edge, relatively narrow round seat tube, teardrop shaped seat stays (sole aero advantage over P2), 650c, carbonaero fork

Assume all else stays the same. Yeah, including position. Cockpit is not very aero (round bullhorns and profile bars). Assume time around 1 hour. Wheels are 650c H3 (clincher). Prorace tires.

Thanks for any ideas you have.

---

________
It doesn't really matter what Phil is saying, the music of his voice is the appropriate soundtrack for a bicycle race. HTupolev