Sorry I don't want to sound like a dick but you didn't climb 2 Aple d'Huez's back to back.
The classic Alpe d'Huez climb is 13km with 1130m of elevation gain at an average of 8%.
I have Alpe d'Huez 1150 m 15.2 km with 7.6% grade …
I also see online 14.5 km/1150 m
http://cyclinginfo.co.uk/blog/procycling/alpe-dhuez-climb/
and 13.2 km/1071 m
http://www.climbbybike.com/climb.asp?qryMountainID=5
So I guess it depends on where exactly you choose the start and finish. Similarly the maximum gradients reported range from 10.2% to 12% depending on where you look.
I agree you have a point that it is not exactly 2XAlpe … but in my defence I'll say the Schelcks didn't have a bone-rattling hardtail descent down 7th Secret after their climb.
Pretty sure the stats put out by your gps are total vertical up and down. In order to break 10,000 ft of climbing on the shore you would need to climb from sea level to the top of the tram on Grouse 3 times. A single Alp d'Huez climb would be from sea level to just below the peak of Grouse.
I'll bet if you walked/cycled from sea-level to the top of Grouse, depending on route, your accumulated elevation change could be well over 4000 ft, with a lot of up-and-down accumulated at the lower elevations.
The GPS stats are for up … the down was also recorded and was about 1.5 m off the up for the whole trip. I uploaded .gpx file to a website that did the plotting/profile/overlay and it reproduced the numbers.
I've calibrated this GPS with a Polar barometric altimeter (considered more accurate and higher resolution) on road rides (so a different situation again - to Whistler and back) and the numbers agree within about 5%. As to the tiny little ups and downs - yes I do wonder how accurate the GPS is with those, but if it is an artifact, it would (?) be a random error where uphills and downhills are exaggerated randomly up or down, due to the GPS cutoff resolution and number of triangulated satellites while on the move. However the data for the loop shows that all uphills and downhills add up exactly over the entire ride. I should check and see if I can apply a smoothing to the curve to more closely reflect altitude resolution.
The amount of paved road ridden was very small - probably about 2.5 km, virtually all of it between the Lynn Creek to Mountain Hwy. Believe me - there is a lot of climbing and descending in all those "flat" parts (eg Bridle Trail - probably of the order of 1000 feet gained and lost traversing there and back with no net elevation change). I've also measured the the amount of climbing on the 7th Secret "descent" with the Polar barometric altimeter, and the "descent" involved a fair amount of uphill, I believe it was in the order of around 1,000 feet.