I guess ultimately we are all going to see what we want to see here, and we have released so much data over the years that you can always go back and nitpick it however you would like. In the end, I welcome any of you to go and create your own data sets and post them for free on this website. You could easily replicate the cancellara test for about $10k plus the cost of the product, travel, lost time, etc...and most of the other data sets for $3-4k each...but remember, that every time you want to try a different tire, or change air pressure, look at additional yaw angles, etc... you are effectively doubling and then tripling the cost and time involved to test
As for the FP data vs. teh other data set, you are talking about two totally different wheels, tested at 2 totally different tunnels (Texas A&M and one at SanDiego), plus different tires, and ultimately the A&M set (the one of hte fp site) uses the A&M tunnel protocol of yawing from 0 to 30, which we do in reverse as a 30 to 0 sweep as yawing the wheel in both directions yields a rather large hysteresis. The 30-0 gives you the worse of the 2 graphs (but the more repeatable one) and more importantly shows the ability of the wheel to perform after stall has been acheived, rather than starting with perfect flow conditions and seeing how long you can get the air to hold on...we feel this is more representative of real performance as real air is actually rather turbulent. However, the biggest issue here is that the FP data set on the FP site is was the prototype wheel for the NEW FP60 which uses 8 fewer spokes and an updated rim shape, while the other data set was taken using what was then the current production FP60...so other than it being a different product at a different tunnel using a different protocol, we are most certainly a bunch of manipulative and lying bastards.
As for our data matching or not matching the Hed data, I can say that their data for the H3 does not match any we have ever created for that wheel with a 22mm tire, and looks more like that wheel with a true 19 or 20mm tire (tires measuring out larger than 20 really start to degrade the performance of that wheel). Our data for the H3 and the 808 matches within a few % of the Tour magazine data which was conducted with 22mm Continental tires. They even replicated exactly our claimed stall angle of the 808 at 12.5 degrees and discuss this in the 2005 wind tunnel test. They retested the 808 in 2007 using smaller yaw increments and matched it again...and again actually posted even lower numbers for the 808 than we have ever published. In reality, it is the Hed data that cannot be replicated for the tire in question, as that data much more resembles the pay per view data which was using a 20mm tire (which truly measures 20mm or less) and at an unrealistically low pressure, and seeing as tire pressure increases tire width and alters tire shape, this is very, very important.
As for the cervelo test, they were using a wider tire in that test and the results were almost exactly as we had discussed in our dicussion on ST about the stinger90 over a year ago, so I'm failing to see what is new here. The fact that the wheels are within a watt or two of each other with that tire says alot for the relavance of the 808 seeing as the 808 concept predates the stinger by almost 2 years, and was designed for 21-22mm tire and not 23 as tested. Not to mention that the 808 has 11% less side force in a cross wind for more controllable handling, and is lighter...but thinking of those things may lead us to have a more nuanced discussion on wheel choice...which is clearly NOT what this discussion is about.
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds! http://www.marginalgainspodcast.cc