@hozn 58432 wrote:
But I was thinking of a power meter. My understanding is that holding ones power constant is something that [power-meter-equipped] cyclists do with some regularity; perhaps it would not be possible to do so with enough fidelity to track subtle bike differences, but I would imagine the differences between a CaBi and a Shiv would bubble above the noise.
But what is trying to maintain constant power really telling you? Isn’t one of the big arguments for different geometries the fact that they change how you apply power, and how much power you need? I guess I’m wondering what the point would be to riding a bunch of bikes in an arbitrarily identical fashion–do you sit bolt upright on the road bike or aero on the cabi? I think it’s a fair guess that a $10k Madone will let a pro racer go faster than a cabi could (the main questions being how much performance each dollar buys you, and how much each component is worth). But what if you can’t bend as much as a pro racer and can’t sit comfortably on the Madone for more than 10 minutes? Or, would the Madone be better or worse for a middle aged guy going on a 1200k than something a bit more relaxed? You’re going to sit differently on those bikes, your power output is going to be different, but one will probably let you accomplish you particular goals better than the other. If, as suggested before, the point is to predict how each bike would work for you, I don’t think your proposed test would get any closer to answering that question, versus riding them yourself. Controlled tests can generate some data, but necessarily provide any conclusions–especially when there are a bunch of uncontrolled variables, as there will be when comparing radically different bikes. The danger (and the problem in the industry) is that people start making leaps with limited data, and start making conclusions which aren’t really supported by the data. Sure, it’s hard and time consuming to try different bikes, but I just don’t think there’s a shortcut that truly gets any closer than a WAG. (Cherry picking data to fit a preconceived notion sounds more scientific than just presenting an unsupported preconceived notion, but isn’t really.) And the bottom line is that even if there’s no scientific reason that one bike is better than another, if it makes you ride better, or even if it’s just more fun, then it is better.
.
