Look at this graph. Thirty-four overweight women (each bar represents one woman) embarked on an exercise program for 8 weeks (about 2.5 hours exercise/week). There was 100% compliance. They were told not to change their diets. Sixteen of them lost some body fat, but 2 had no change, and 16 actually gained body fat. If you looked at the group as a whole, there was a slight increase in body fat.¹

Fat gain happened because their bodies expended less energy during the rest of the day, both metabolically (not consciously) and behaviorally (say, taking the elevator instead of the stairs). It didn’t happen because they ate more, although that was the reason in other studies, e.g. Individual variability following 12 weeks of supervised exercise: identification and characterization of compensation for exercise-induced weight loss.

The New York Times just described this same scenario in a new study.²

Exercising but Gaining Weight, New York Times Well Blog, 12 November 2014

Scientists at Arizona State University in Phoenix recruited 81 healthy but sedentary adult women. All of the women were overweight. None had exercised regularly in the past year.

The women were told that they would be joining a fitness study and would exercise in order to improve their aerobic endurance. The scientists asked the women not to change their eating habits in any way.

The women walked on treadmills at the laboratory three times per week for 30 minutes at a pace that represented about 80 percent of their maximum endurance.

At the end of 12 weeks, the women were all significantly more aerobically fit than they had been at the start. But many were fatter. Almost 70 percent of the women had added at least some fat mass during the program, and several had gained as much as 10 pounds, most of which was from fat, not added muscle.

Those women who gained weight and body fat had compensated for the energy they expended in exercise, either by eating more or moving less. As all of these studies indicate, some people are more prone to this compensation than others. I do know that people with type 2 diabetes are, metabolically, more resistant to loss of body fat from exercise than healthier people.

I’ve learned a few things about exercise over the years. One – there is an amount that is required for good heath. Two – too much can be detrimental. Three – if the goal is purely weight loss, changing what you eat will likely be more effective than increasing your activity.

¹ Behavioral Compensatory Adjustments To Exercise Training In Overweight Women, Medicine and Science in Sports and Exercise, June 2010
² Predictors Of Fat Mass Changes In Response To Aerobic Exercise Training In Women, The Journal of Strength and Conditioning Research, October 2014
Resistance To Exercise-Induced Weight Loss: Compensatory Behavioral Adaptations, Medicine and Science in Sports and Exercise, August 2014