Something that has bothered me from my field tests and general observation over time is that I haven't seen the sort of advantage from disk wheels that I expected. In several instances I have pre-rode TT courses with training wheels and seen little or no time improvement in the actual race with a disk on - despite having the same (or higher) power. Obviously, wind and other environmental factors play a role -- it's just troubling that after a number of times, I have yet to observe a significant advantage.
Going back to published literature on this point I see some confirmation:
* In JC and JM's article in High Performance Cycling, the advantage between a "aero rim" (Hed Alps is depicted in the article) and a "disk rear and composite front" is only 11 seconds over 40km - and that assumes a flat course. I would imagine some, if not all, that 11 seconds comes from the composite front.
* In the German Triathlon mag, a composite spoked wheel (xentis) outperformed the disk combo in a field test on a track
* In a subsequent article, the Zipp 999 ranked below 4 different spoked or composite wheelsets, including Zipp's 808 pair
* Rennrad showed similar results in their field tests (I can't seem to find that one at the moment)
Of course, the tunnel tests show disks having an advantage at higher yaw angles - like >15-20%. But that suggests the peculiar irony that a disk might not help you unless you're really slow, or you're climbing a big hill.
I found the comments from Andy from Hed (in a thread on this site) intriguing -- particularly the suggestion that disks are not as effective in an aero frame as they are sitting in a stand all by themselves in a wind tunnel. It squares somewhat with my experience and helps explain LA's enigmatic choice of tri-spokes for last year's TdF TT.
-jens
My latest book: "Out of the Melting Pot, Into the Fire" is on sale on Amazon and at other online and local booksellers