It makes sense that this would work. When I log my foods into fitday, and I see the total calories for the day, it doesn't mean as much to me as what the report for the week, or two weeks, or month. It's how the total daily calories average out.
Interesting way of looking at the average: If you eat 2000 calories one day and 1000 calories the next, then your average for those two days is 2000 + 1000 = 3000. 3000/2 (days) is 1500 calories. That's your average calories for those two days. You would have gotten the same average if you had eaten 1500 calories both days! 1500+1500=3000 calories. 3000/2 = 1500 calories. The average, 1500 calories, is the same if you ate that number of calories both days, as when you ate 2000 calories one day and 1000 calories the next. Cool!
It sounds simple and mundane. But, if you OVEREAT (have a binge day or weekend), you bring the average calories for each day - in a week, say - UP. All that careful dieting and it was like adding food every day. On the other hand, it speaks to the power of really cutting down on calories one day. It's as if you ate fewer calories each day of the week, without actually having to do that!
That's why I don't really like the idea of having 'cheat days' if I'm really mildly cutting calories. If I'm cutting calories a lot, every day of the week, a cheat day is going to impact the average, but I might be willing to live with that... it's not a whole, whole lot of impact. But if I've been kind of rolling along, eating less, but not a whole lot fewer calories, a 'binge' is going to undo even my mild efforts by slightly raising my average for each day. Easy enough to fix, given more time (and more time added to the average) - just 'get back on the horse' and really cut down on calories the next day and see if I can try to 'fix' the damage.
Easy enough to see when you run reports and try the different time periods (week, two weeks, four weeks, etc.)