wrote:

Well, the scale says I lost a pound yesterday.
I know that can't be right, but I'm recording it anyway. :-)

If, after many data points, the actual weight loss ends up being more
than the theoretical, I guess that means I'll be able to calculate my
actual BMR, which must be higher than the 1800 estimate.

Can you tell I used to be an engineer?

Seeing you 1800 cal estimate got my thinking.
I'm using s/w on my PDA to track my eating + exercise. Early on, it
calculated my BMR as 2900 calories/day. Putting the same figures into
http://www.bmi-calculator.net/bmr-calculator/ I get 2500. I guess this
means I've been mislead into how well I'm doing
I also noticed when I got my new exercise bike that it's computer gives
MUCH lower calorific expenditure readings for comparable effort/time to
my s/w and my previous bike. FWIW, I always log the lowest reading, but
I'm losing a little faith inthe application. Still, as long as the
weight comes off, I guess I don't really mind. I'm intending to
exercise more than I eat for the next few weeks anyway so all my BMR
will come from fat!