You're close. Attempting to do so while not having any understanding of why things happen is a recipe for disaster.
It must be reasoned that obesity is a disorder caused by something more then just 'poor eating habits'. There are physiotypes that will simply resist obesity on the worst of diets. I don't need to link a paper for this, this should be common sense from anyone's experience around enough people.
So, if you reason that obesity is a disorder or sickness... you would then try to treat the cause, not the symptom. In the same sense that a person with a peanut allergy would simply avoid peanuts, a person with obesity may have this state caused by disturbed leptin/insulin signaling. A diet must then be designed to compensate for that disorder. (we know that once surplus white adipose tissue is grown, there's no going back, a modified diet must be adapted for life or those buckets just start filling again)