I'm not anal enough to want exact numbers.
Calorie counting is a rough estimate for me. Likewise for burned calories.
This method has worked well.
I use 20 Calories a mile for easy efforts, 30 moderate, and 40 for superhuman going to die type efforts.
I'm sure it is all BS but my weight is stable or drops depending on my goal.

I also do not bother tracking during days that have huge amounts of exercise and the day after. Ex a 120 mile ride with 13,000 feet of climbing.

The only reason I track is to keep me from eating everything in sight due to my insatiable appetite.

I live in Cleveland also...which big hill are you going down that is broken up with lights or sharp curves. Also, what hill do you find the most fun to bomb down, I'm looking for raw speed here.

For raw speed - Vally Parkway just South of the Turnpike near York or is it State Road. Also behind Sleepy Hollow Golf Course, the hills down to the river.
Can't think of too many in Cleveland that are a problem for bikes. However we did an Inline race near Vally City with hills and at the bottom of every hill was a nasty turn.

A good example would be, Wednesday's TdF decent into Gap. Anyone not getting paid big money, would be on their brakes all the way down the mountain.

Here is a simple thought experiment. According to calculations derived from your power meter you say you are "burning" 500 calories per hour. Sit on your bike in the garage all day without turning the cranks. How many calories does the meter say you've burned? Zero. Does that sound right?

My point here is that a lot of people misinterpret the data they get from power meters. You burned 500 calories per hour generating the wattage measured by the power meter. But your body has an additional base metabolism that also has to be accounted for just to keep you alive. In other words your body is burning calories doing a lot of things that your power meter does not measure. So the power you produce won't equal the calories you burn; it is only one component.

Back to the thought experiment, you'd probably have "burned" 2400 calories just sitting there all day (YMMV). Translated to your own experience that would be an additional 100 calories per hour on top of the 500 you measured which is much more in line with many of the estimates you see from various sources.

Yes, it is how much work you have done over your base metabolism. Your "experiment" is flawed because those estimators on dieting sites and the like take that information and add it in as activity calories burned, without accounting for your base metabolism. So if they are including your sitting around doing nothing calories, then you end up double counting. And if they don't, they are just over-estimating. Unless the calculator says what it is assuming, you can't really know. But you know with a power meter that it is only measuring the work you actually do. Makes things much cleared IMO.

So with all the over calculation of the calories used, why would a company like Garmin [and others] come up with some over inflated number. You would think that with all the money they have with R&D that they could come up with a better number. Does anyone know how they come up with those numbers/calculations? I know as a Mercedes technician that the Magic 8 Ball has worked wonders for me. The amount of customers that come in with intermittent issues that get resolved with this thing is amazing. Maybe they have an industrial strength version.

That's what my Bell F20 (yes I know it's no power meter) says when I ride at medium to high intensity for an hour...I usually do 15 miles in that hour and get about 450 calories burned...at about 17-20mph.

How can you ride 15 miles in an hour at 17-20mph?

Ops, forgot what my computer read after that particular "ride". I did that on my trainer. But on a ride from my house to Venice beach's bikepath..headed north to the end of the paths near Malibu..then rode back home. His iPhone app said we burned 1400 calories in 2 1/2 hours of pretty high intense riding...

I was thinking autopause.