How to Choose Between Diets Without Joining a Camp

The dietary debate is conducted as though one pattern must be right for everyone, which is the least likely of all the possibilities. A more useful question is which problem you are trying to solve.

Start by naming the variable

• If fasting insulin or triglycerides are high, or waist circumference is the concern, carbohydrate is the variable — a ketogenic or low-carbohydrate pattern moves those fastest
• If the symptoms are joint pain, skin, or gut, the variable is more likely a specific food, and an elimination approach answers it where macro-counting will not
• If nothing is obviously wrong and you want a pattern with long observational support, the Mediterranean pattern as actually studied is the best-evidenced default
• If you have never removed ultra-processed food, that is the variable, and no further choice is needed yet

What every workable answer has in common

Sugar, refined flour, industrial seed oil, liquid calories and ultra-processed food are absent from all of them. If a pattern you are considering keeps any of those, it is not really in the conversation, whatever it is called.

Adherence is a nutrient. A slightly worse diet you will still be eating in a year beats an optimal one you abandon in March.

How to test rather than believe

Pick one pattern, run it properly for eight weeks, and measure something at both ends: waist, weight, blood pressure, and if you can get them, fasting insulin and the triglyceride-to-HDL ratio. Those two blood markers tell you more about whether it is working than the scale does, and they move on a timescale you can act on.

If the markers improved and you feel well, you have your answer. If they did not, change one variable rather than switching camps — the most common reason a reasonable diet fails is that it was not actually being followed.

A note on the arguing

The people arguing hardest online tend to have solved their own problem and generalised. That is not dishonesty; it is what a dramatic personal result does to anyone. It is also why the sensible move is to treat every strong recommendation, including the ones on this site, as a hypothesis you are going to test on yourself rather than a conclusion you are being handed.