[falcolas]

> The answer is if you want to lose weight it doesn't matter. You just need to average a deficit over time.

If this were the case, Weight Watchers (which is at its core a calorie limiting system) should have a near 100% success rate. It doesn't. A casual perusal of Google shows closer to a 50% success rate for a 5% body weight change.

I can't think of many medical treatments where we go "yeah, a coin toss level of success is good enough for me."

As for the accuracy of the measurements - it does matter. Healthy weight loss rates are, mathematically, in the 10% calorie reduction range. That's less than the error rate for the caloric labeling of many foods.

[6cd6beb]

>If this were the case, Weight Watchers (which is at its core a calorie limiting system) should have a near 100% success rate. It doesn't. A casual perusal of Google shows closer to a 50% success rate for a 5% body weight change.

First off - that's an astonishing success rate against a backdrop of an article saying, literally, surgery is the only reliable option.

Second off - You're assuming 100% compliance. I know people with personal trainers who are gaining weight because they don't stick to the program. I halfheartedly went through a weightlifting routine for years and got nowhere because I never really stuck to the program. 100% compliance in the field of weight loss is just not going to happen.

[sohex]

I don't think you can draw meaningful conclusions from a self-enforced system like Weight Watchers either though. Self reporting particularly when it comes to anything diet related is going to be heavily skewed. There will always be social pressures within a group like that to over report exercise and under report intake.

[falcolas]

If you can't trust the results of a self-selected group of individuals who are paying a good chunk of cash for the express purpose of loosing weight... who the hell can you trust?

[sohex]

Generally speaking you're absolutely right, but with diet the social pressures are significant enough to compromise analysis even if the group is paying for it. There have been quite a few studies on the topic, e.g.: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.103...