Mr Haematocrit
msg me on kik for android
- Location
- Out of the saddle
Well, wind tunnel data is as always a bit of an odd one too, however, at least in a tunnel the results are generally repeatable in context, in the test you quoted, the noise would be greater than the actual performance benefit of one frame over another, looking at how close some of the lap times were. Also, variance in the timekeepers reaction time would account for a few fractions of a second alone.
I would have suggested a fairer way to test would be to set them all up the same, with the same kit, have a rider ride round a velodrome with a power meter attached and compare the CdA data (this way deviation from the same line wouldn't matter) and you would get a good estimation of which bike is quickest at low yaw (then again though, if you get a few that are close, the noise in the data is likely higher in magnitude than the actual differences you are trying to detect, i.e. in the real world, it's neither here nor there in the context in which these bikes are to be used really). Of course you can do the same on the road using a circuit or out and back course, but it would be less repeatable due to wind, surfaces etc.
I would dare to day that what you suggest would also be subjective in terms of determining the bikes performance. I have always found the Venge for example to be far stronger and stable under braking when descending than the Pinerello Dogma which is an important part of the overall performance, but this is not going to be repeatable in a velodrome either...Interesting problem and I'm not sure there is a solution with regards to testing which is not subjective.