- Mar 10, 2009
- 2,973
- 5
- 11,485
Re:
1. It's pretty clear there are still a lot of people about with very little understanding of HR response, let alone the manner in which such data is collected.
2. There is even less understanding about the nature of power meters and the manner in which such data is collected and the factors that affect both a rider's power output and the accuracy of devices used to record it.
Even if the SRM was perfectly calibrated and torque zero was spot on, these non-circular rings introduce a variable error into power numbers, and generally over report power by ~ 2-5% but the error varies depending on the nature of riding. e.g. in general the lower the inertial load, the greater the error. Other factors include pedal rate and the torque profile of the rider (which is also multi-factoral).
For instance on hill climbs the error is typically larger than on flatter roads. And the error is also greater when accelerating and with different torque profiles, e.g. standing versus seated.
This is due the assumption of non-variable crank rotational velocity during a pedal stroke used by such power meters when generating power data. That assumption is violated to a significant extent with non-circular chainrings and power data is typically artificially inflated, the error is not constant but is variable, and the amount or error introduced depends on various factors including those already mentioned.
If "experts" are going to analyse such data, I would expect them to at least have a grasp of such power meter basics.
3. Even if the data were perfect, it still doesn't say much about doping status. Despite that, it's evident that many will persist with their logically fallacious confirmation bias (whichever way they lean on the subject of a rider's doping status).
4. Suggesting this data is confirmation of the accuracy of a power estimate made from climbing speed is ignoring the quite possible bias error in both estimates due to unaccounted for errors. Two data wrongs don't make a right. I'd expect better from scientists.
5. I don't care whose data it was, stealing and theft is not cool and is wrong. Condoning it is poor form. I'd be pretty pissed off if any of my or my client's data was stolen, or was used without permission.
6. As for Vayer, he can be mathematically challenged at times and doesn't let the facts get in the way of generating publicity. I'd suggest looking to someone far more credible when it came to estimates of and/or analysis of power data. Becoming a thief further diminishes one's credibility, although I have no idea who was responsible for such theft. Knowingly using stolen data is still piss poor.
danielovichdk2 said:
red_flanders said:
IzzyStradlin said:
1. It's pretty clear there are still a lot of people about with very little understanding of HR response, let alone the manner in which such data is collected.
2. There is even less understanding about the nature of power meters and the manner in which such data is collected and the factors that affect both a rider's power output and the accuracy of devices used to record it.
Even if the SRM was perfectly calibrated and torque zero was spot on, these non-circular rings introduce a variable error into power numbers, and generally over report power by ~ 2-5% but the error varies depending on the nature of riding. e.g. in general the lower the inertial load, the greater the error. Other factors include pedal rate and the torque profile of the rider (which is also multi-factoral).
For instance on hill climbs the error is typically larger than on flatter roads. And the error is also greater when accelerating and with different torque profiles, e.g. standing versus seated.
This is due the assumption of non-variable crank rotational velocity during a pedal stroke used by such power meters when generating power data. That assumption is violated to a significant extent with non-circular chainrings and power data is typically artificially inflated, the error is not constant but is variable, and the amount or error introduced depends on various factors including those already mentioned.
If "experts" are going to analyse such data, I would expect them to at least have a grasp of such power meter basics.
3. Even if the data were perfect, it still doesn't say much about doping status. Despite that, it's evident that many will persist with their logically fallacious confirmation bias (whichever way they lean on the subject of a rider's doping status).
4. Suggesting this data is confirmation of the accuracy of a power estimate made from climbing speed is ignoring the quite possible bias error in both estimates due to unaccounted for errors. Two data wrongs don't make a right. I'd expect better from scientists.
5. I don't care whose data it was, stealing and theft is not cool and is wrong. Condoning it is poor form. I'd be pretty pissed off if any of my or my client's data was stolen, or was used without permission.
6. As for Vayer, he can be mathematically challenged at times and doesn't let the facts get in the way of generating publicity. I'd suggest looking to someone far more credible when it came to estimates of and/or analysis of power data. Becoming a thief further diminishes one's credibility, although I have no idea who was responsible for such theft. Knowingly using stolen data is still piss poor.
Facebook
Twitter
Instagram