Everyone in the study was on a diet, but they put some people on the "wrong" diet for their genome. They were the control. The 2-3 times more weight lost was comparing those two groups in the study, not comparing to people outside of the study.