You weight loss will be (on average) about one pound for every 3500 calories of energy deficit that your body has. Even if you burn 3500 calories running, if you spend the rest of the week hoovering onion rings and cheesecake, it’s not going to do much.
FWIW the explanation of the discrepancy is given here.
Of course the amount of fat lost due to your physical activity (or even weight lost) is not really a straight up linear relationship to the amount of calories you burn anyway. Other factors come into play such as how much your anabolic needs are from the activity you did (rebuilding from wear and tear even without putting on new muscle mass, putting on muscle mass, vs catabolizing muscle mass), whether or not your baseline metabolic rate is changing as a result of your activity, etc. The 1000 calorie deficit per day (of increased activity and intakes changes) leads to the ideal target of 2#/week is a great rough guide, but anyone who expects the scale to play along in a nice linear fashion because that’s how the calculations worked out may be disappointed and or pleasantly surprised by what the scale tells them any particular week, even as they are making great consistent progress independent of what the scale announces.
That’s brilliant, exactly what I wanted, thanks. I’m just going to quote it here in case one day someone else happens by with the same question and the link is dead.