Well, if you say BS to that then I suggest you stop crying for information and go out and generate it yourself. We've established the information you are seeking is not in the public domain. Otherwise, this is just a circle jerk. The status quo is the status quo; it has no need to defend itself.

BTW, if you don't want the state of the art frame with its price, most manufacturers offer a second and third tier.

Well, they do have their different agendas... ;)

Industry wants to sell stuff. University wants to publish. Both endeavors include their share of human factors.

Does anyone reckon a too-stiff bike can feel slower and less responsive?

Here's an anecdote: :p

I have a 58cm, old CAAD8, which I reckon is quite stiff. I got the frame cheap on Ebay without a fork, so I threw on a cheap HASA (Scott copy) fork. The bike felt fantastic. After a few weeks, I decided to beef it up by putting on my stiffest fork, which is an old Columbus Link (stiff, chunky blades and an aluminium steerer).

Well, the bike did feel stiffer, but slower!! :D It obviously lost some of its liveliness and felt a little less responsive. I naturally put the softer fork back on.

Incidentally, the CAAD is great (uncany! :)). If, around ten or so years ago, I went from riding my old 531 steel to test riding a CAAD, I would've emptied my bank account to buy one.

So there! Good story? Darn tootin it is. :p

I don't think it is even about proving one position or the other, but rather just knowing what's what.

True, not in the public domain...but why does that translate to never will be? The status quo is the status quo until it's not any more. It has no need to defend itself, but neither does anyone else have any need to defend it. The status quo, when challenged, either survives or gives way. This post is about letting the bike makers know it is time to give way on this. As for second and third tier bikes, they might be the exactly right choice, but how will we know without the data? Oh hell, we are right back to the same place.

Well, if you want to do it right, do it yourself, right? The reason why you aren't is the same reason why the data isn't in the public domain.

Bike manufacturers and several magazines have no problems publishing extensive details about drag savings due to aerodynamic improvements, and comparing them to competitors' bikes. E.g. @ 40kph, a typical aero frame might save 15-20 watts vs a round-tube frame, under ideal conditions.

If a Giant Propel was 5% faster than a Cervelo S5 in a sprint, they'd be shouting it from the rooftops, not hiding it.

Plus, if Giant can develop a methodology to figure this out for their frames, they could easily run those tests on competitor's frames.

The idea that a quantification of the power benefits of a stiffer frame is a trade secret is mildly amusing.


The simple fact remains: The public has no evidence to back up the claim that "stiffer = better power transfer". It may be true. It might be true only in specific situations. It could be false. We just don't know.

There is little basis for a high degree of certainty when you don't have access to the evidence.

I don't see why this is the case. I'd think that aerodynamics is much more difficult to analyze than power transfer.

An experiment wouldn't be hard to design. Get a set of reference wheels, pedal-based power meters, and a hub-based power meter. Calibrate the meters. Do a large number of runs on a trainer at various wattages, with different frames and a few test subjects, where the only change is the frame. Compare the differences between the pedal-based and hub-based meters.

Sounds a heck of a lot easier than building a wind tunnel.

Thanks to all who helped us make it this far.

Our school (Engineering and Applied Sciences) now gets more funding from industry than from government grants, which, if you ask me, is a nice mix. The industry projects tend to be focused on development, and give them access to specialized equipment and dare I say certain expertise. The governement grants tend to be more high-risk, fundamental research, which can be really fun because of the intellectual challenges. But it's funny how the tenure-track faculty (I'm NTT) talk about their research these days: from day one it's about getting tenure; the actual research is almost an afterthought.

Who is willing to pay a qualified person to do all the work?

Why no published data? My educated guess is that the "crowd-sourced" consensus is probably right - stiffer is better - but not as right as they think they are. The incremental improvements at this point are probably insignificant, so for marketing purposes it's better to call it "stiffer" than to put out unimpressive performance metrics.

It's still possible that the conventional wisdom is wrong, and that's why we don't see research and measurements. That possibility is what makes the questions interesting IMO.

Since you can do the entire experiment in a stationary environment you can build a bike that would effectively be ABSURDLY stiff. Simply add horizontal/vertical stiffeners to the BB connected to a stationary frame. A gorilla could stand on the pedals and the BB bracket wouldn't move.
I'm guessing the truth is Giant is right, beyond a certain point, frame stiffness is irrelevant.

Another way to look at it. The easiest way to stiffen a frame is by adding weight. Since race bikes have a minimum weight, if stiffer is strictly better, you'd see frame weights increasing to boost stiffness. The opposite is happening, frame weight is still going down. Design is being used to reduce weight while effectively maintaining stiffness.

I couldn't have said (any of) it better!

I don't see how it could be likely at all. Given a specified force input that causes deformation, a stiffer frame will absorb less energy while trasmitting more to where it's "supposed" to go.

So, for a flexier frame to be as efficient as a stiffer frame, the flexier frame would have to return a much higher percentage of energy absorbed in flexing to the drivetrain somehow. I can't see how that could be more efficient than simply transmitting that energy directly to the drive train in the first place.

As has been pointed out a few times, the dominant theory is that the frame acts like a spring. When the spring is released, that force winds up going to the pedal on the upstroke. The only loss would be to friction, and it's a very small amount.

And this is why the answer cannot be settled without data. If we don't actually know what's going on, we can never get beyond mere conjecture.

All reasonable, but it is not about what we think. It is about what it is, i.e. what the data says. Everybody has an opinion and not one of them makes any difference. As Joe Friday used to say on "Dragnet", the facts ma'am, just the facts." We just don't know the magnitude of the effect at today's levels of stiffness. That is what I want to see, and thanks to folks for their opinions, but it is not the same thing. It is quite possible to be technically right here (stiff is better), and yet for it not to make any important real world difference. The effects may be very small. Or not. Who knows?

I actually don't believe much of what I call misdirected effort does come back. But then I don't believe there is all that much of it to begin with and the differences in modern frames is trivial and getting smaller all the time. You and I disagree on this, but agree on the main question, how will we know without the data. Do you think we will make it to 10?

Past that, your pedaling motion applies X force laterally into the BB. Frame stiffness determines how far that force moves the BB, which may feel better, but you're applying the same force, and either losing it or having it returned.

I'm shocked this is still going with the same arguments being repeated.

I lean towards agreeing with you, but for the bolded part the question is why? It's not immediately obvious that you're putting in more energy into bending the flexy frame a lot as opposed to bending the stiffer one a lesser amount. Even hypothetically resolving that I have too many questions about the rest of the system. What happens to the frame's position and motion in each case, and how does that affect the rider, the efficiency of his efforts? What changes in the tire/road, with the differing frame angles due to different flex? Moving the bike back and forth more, how much energy is expended? Other, even seemingly sillier considerations. Maybe everything there is trivial, but it's not clear that it necessarily is.

Welcome to "The 41." ;)

I'm just hoping that someone will find actual evidence that has been overlooked.

Anyone see the way Cavendish's bike was bouncing around on the cobbles today in the TDF?

Given the same force, since energy is force times distance, the frame that flexes the most will absorb the most energy.

Both frames flex some and absorb energy, but since the flexier frame absorbs more energy, for a flexier frame to be more efficient than a stiffer frame, the flexier frame would have to return a higher percentage of the energy absorbed in the flexing.

Because no transfer of energy is ever lossless.

For a flexy frame to even be as efficient as a stiffer frame, the flexy frame would have to perform TWO energy transfers of MORE energy AND do BOTH transfers MORE efficiently.

And that's just to MATCH a stiffer frame's efficiency.

Figure the odds.

Yeah.

Not damn likely at all.

Aerodynamics is downright easy - put the thing in a wind tunnel, blow air over it, measure the static force. Think "sideways scale".

No, the theory is that the more compliant frame works like a spring, and almost all of the energy is returned in what happens to be a useful fashion.

The only energy is lost is via friction, which is very small in this condition.