I suppose the upper panel is the velocity measured with a "wheel magnet sensor". When the velocity goes up, power goes up too ! However CdA can come down, for example by a change in body position. It is difficult to conclude from a screen shot without more background information.

Without knowledge of the context, it could mean all sorts of things or it might mean nothing. All the twitter link says is it's an example of live data being recorded. It says nothing about what was going on.

It may simply have been a rider riding around the track with no specific instruction. Perhaps the rider started out sitting up a bit before beginning a harder efforts and changing bike position.Maybe there was another rider on track and this was just looking at one rider's data while riding around. Perhaps that was simply a demo of the real time wireless data capture system and wasn't related to specific aero testing protocols. Maybe it was just a demo/teching/learning exercise. Who knows? The tech is used for more than just aero testing, although that's its best known application.

It's a bit of a stretch to come to any conclusion about that screen shot without context of what was going on while that data was being recorded.

Actually, I think that even without the context of the test, it shows how well the system works. You can see that the variance around the estimate is quite low. This is a very sensitive technique that can pick up small differences. You're right that without the missing context we can't tell what those differences are but the screen shot shows good precision, so you know the differences are systematic and not random.

Can you recognize the division of CDA axis? Do you know how strong the data smoothing was? (I am pretty sure the data is smoothed - intentionally or not - otherwise CDA should vary with the cadence)

From other, better screen shots you can find in the internet I would estimate a CDA variation (one sigma) of around 0.005 m². This about the same delta CDA I currently get with my software from round to round and a cyclist who can hold the black line.

See: https://www.flickr.com/...933@N05/15318591055/

I don't know how much smoothing there is in the data, but the data are probably collected from ANT+ sensors which transmit at 4Hz. That said, most PMs that use a reed switch only update the power signal once per revolution so power gets updated less frequently than that; and similarly with speed. However, one of the strengths of this method is that there's "natural" smoothing from inertial mass. That's something that gets ignored in most analytical models which presume that the observations are iid. I think in one trial I was getting a sd on CdA of around an order of magnitude smaller than that (by memory, ~0.0005 m^2).

I agree lots of opportunity for those with a dog in the hunt to provide knowledge of the context.

---

=================
Kraig Willett
http://www.biketechreview.com - check out our reduced report pricing
=================

it looks like the vertical ticks for CxA are every 0.010 m^2 to me. First labelled one we can see is 0.21 and the highest is 0.26. this is consistent with the reported CxA's in the ovals to the right as well (values there seem to vary from 0.234 to 0.224).


I can't disagree on that one. on an instantaneous basis, it seems to be a function of power, also. i wonder what the sigma is across days. that's a tough piece of work from an experimental design perspective - or not! ;-) even without context on this screenshot, it still suggests more power equals lower CxA, IMO.


very cool! CxA of 0.19X. what is your height/weight and shoulder/hip width while riding in that position?

these are sweet: https://www.flickr.com/...7933@N05/11052664913

---

=================
Kraig Willett
http://www.biketechreview.com - check out our reduced report pricing
=================

 


This is from an arbitrary run from a 175cm tall but skinny girl (~55kg, in the off season).

Never really measured shoulder/hip width, but from my experience with aero testing people I can state her body is made for racing a time trial machine! Just the right proportions together with high flexibility coming from a ballet background. I have the same hight and ~62kg, but I am not even close to her concerning CDA. (I prefer crits and mountains)

or you can be 175cm and 79kg and come in at 0.200 ;)

What size of frameset are you riding?

Large with low/far stem