Probably all of us are guilty of that at some time or another. I know I have been and probably will suffer episodes of it in the future.

If a person's persistent about it, particularly in expecting number from one ride to match the numbers on another ride, particularly Calorie Count, then I'll accuse you of being a bean counter.

As for exactly how any particular Garmin device or other manufacturer calculate their numbers and averages, we just don't know for certain. Much of that kind of detail in what factors go into calculation is going to be considered proprietary information.

Sometime you can glean some useful hints over at the Garmin Forums either a subforum for that paticular device or one of the devoloper subforums. But not regularly. You just have to stay up to date on the posts and hope to stumble across a post that seems as if someone had some true insight by being or knowing a Garmin developer, was good at reasoning and testing their hypothesis to at least make it a good theory. Or simply is a great BS'er.