Old 08-25-13, 02:23 PM
  #21  
brianogilvie 
Commuter & cyclotourist
 
brianogilvie's Avatar
 
Join Date: Feb 2009
Location: Hadley, MA, USA
Posts: 496

Bikes: Boulder All Road, Surly Long Haul Trucker, Bike Friday New World Tourist, Breezer Uptown 8, Bike Friday Express Tikit, Trek MultiTrack 730 (Problem? No, I don't have a problem)

Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Likes: 0
Liked 0 Times in 0 Posts
I suspect that you're overestimating calories burned. Bike Calculator can give you a sanity check (when the site is accessible, that is!). My Edge 800 tells me that I burn about 20 Cal/mi when toodling along at 11 mph with my wife, about 30 Cal/mi at 14-15 mph, and around 40 Cal/mi at 17 mph or when climbing serious hills. Its estimate for my D2R2 100K yesterday, 65 miles and over 7000 feet of climbing, was 3200 Cal, or under 50 Cal/mi.

To put that in context, most online calorie meters read 50-100% more. The fact that they all agree is not very useful, because most of them rely on one version or another of the Compendium of Physical Activities. They're generally not independent of one another, so their agreement simply reflects their use of a common source.

You may be underestimating calories consumed, too. I'm not casting aspersions; it's normal to do so, as countless studies with doubly-labeled water have demonstrated. Unless you weigh everything you eat with a food scale, you may be missing some.

However, I wouldn't suggest obsessing about the accuracy of your figures. Rather, I suggest you use feedback, as John Walker suggests in his excellent (and free) ebook, The Hacker's Diet. You have a number for calories consumed, and you know that your weight is stable at that number, meaning that it effectively represents your current TDEE (total daily energy expenditure). A quick and dirty engineering solution to that would be to cut 500 or 750 from that number, while keeping everything else constant, and see what happens. Point is, it doesn't matter so much if your numbers are accurate as long as they are reasonably precise (i.e., consistent with one another), and you use feedback to adjust them.
__________________
--
Brian Ogilvie, Hadley, MA, USA
brianogilvie is offline