@ Dave
Nobody? If there were only exceptions there wouldn't be so much concern about an obesity epidemic. In 2009, about 27% of men and 26% of women in Scotland were obese (and one third of children were overweight).
You only get to be called obese when your BMI goes over 30. A thirty year old man, 5' 8" and 15 stones is obese (only just). He's not huge - not one of those cartoon roly poly fat people used to give a convenient illustration of obesity on TV, he wouldn't get a programme on Channel 5, but he's obese nonetheless.
He needs to lose a pound of fat a week (although his body will fight him over it and conventional dieting will take as much lean tissue as fat). He only needs to lose 30lbs to bring his BMI under 25 to not be overweight but he needs to do it in a sustainable way before Type 2 diabetes kicks in.
I think it's worth getting a grasp of the numbers and the effort involved because when you reach the stage that two-thirds of the population is overweight and one-third is obese you need to stop thinking in terms of individual problems of excessive pie consumption and individual solutions of marginal increases in activity and start looking at system-wide, societal problems and solutions. See below.
@ ruggtomcat
There's plenty of junk science around but the calorie is a pretty reliable measure of the energy content of foods. The body would only be limited in its ability to extract energy if it were unable to break down the food and that is taken into account in current methods and has always formed part of the assessment, right back to some of the original accounts published in the late 1800s.
But there was an interesting review of the issue in The Lancet recently that was more good science. One part of the series was a model that builds in changes to metabolic rates and, of course, the lower number of calories needed to sustain a person as they lose weight, which, paradoxically, suggests that the problem of weight loss is greater than commonly understood.