Do You Have to Cut Carbs to Lose Fat?

The trial tracked 609 adults for 12 months. Half ate healthy low-fat. Half ate healthy low-carb. Both groups got the same coaching, the same support, the same access to quality food. After a full year, the between-diet difference was 0.70 kg — roughly a jar of peanut butter.

Not statistically significant. Not clinically meaningful. Not even close to the metabolic revolution the funders had staked their reputation on.

The organization behind the funding, NuSI, had been co-founded by Gary Taubes and Peter Attia specifically to prove that carbohydrate restriction was metabolically superior. When their own trial couldn't make that case, NuSI effectively dissolved. The people with the strongest possible agenda watched their hypothesis fail its own test.

That $8.2 million also bought something else: every participant was genotyped to find a match between genetics and diet response. The idea was that even if low-carb didn't win on average, certain people might be genetically wired to lose more on low-carb.

The genotype test found no connection. The insulin test found even less. The most expensive personalized-diet experiment in history found nothing to personalize.

But That's One Study

Fair. So what happens when you zoom out?

A separate analysis pooled 19 trials where both groups ate the exact same number of calories — only the macro split changed. The difference: 0.74 kg. Not significant.

A two-year trial tried the same question from a different angle: 811 adults, four completely different eating patterns. Weight loss tracked calorie reduction and adherence across every one. The macros didn't predict who lost more.

And the largest pooled analysis available — 53 RCTs, roughly 68,000 participants — did find a statistically significant result. Low-carb won by 1.15 kg over a full year. The researchers who found it called it "clinically insignificant." That's about the weight of a water bottle.

Four different research designs. Free-living, calorie-matched, calorie-targeted, and broad meta-analytic. Different labs, different years, different continents. Every one arrived at the same place: the macro split is not the lever.

If you came here hoping the evidence was mixed, it's not. If you came hoping one of these studies had a flaw that the others didn't — each one's limitation is covered by another's strength. The convergence leaves no methodological refuge for the belief that cutting carbs is required.

What About Metabolism? What About Insulin?

You've probably heard that low-carb makes you burn more calories. One study did find a metabolic-rate increase of 209 to 278 kcal per day on a low-carb diet. But that finding came from a specific context: participants who had already lost 10% of their body weight and were in a maintenance phase, not actively cutting.

Whether that metabolic edge exists during the deficit phase is a different question — and in every trial that measured actual fat-loss outcomes over 12 months, it didn't translate into more fat lost. We examine that metabolic-advantage dispute in depth separately.

And the insulin argument — the idea that carbs spike insulin, which drives fat storage — was tested directly under strict metabolic ward conditions. The hunger and intake predictions of the carbohydrate-insulin model did not hold up.

The theoretical engine behind "carbs make you fat" stalled when it was actually measured. That mechanism question gets its own deep analysis elsewhere in this series.

The Part Nobody Talks About

Here's what should have been the headline of every diet debate for the past decade.

Within each diet group in the largest trial, person-to-person weight loss ranged by approximately 40 kg. Some people lost 30 kg. Others gained 10. Same diet. Same duration. Same researchers. Same coaching.

That individual variation was 57 times larger than the difference between diets.

Both groups spontaneously cut about 500 to 600 calories per day — without being told to — simply because they focused on food quality instead of counting anything. The two-year trial with 811 adults confirmed the same pattern: what predicted results was how consistently people showed up, not which macros they chose.

The question was never "which diet." It was always "which diet can you actually sustain."

The Friend Answer

Based on everything we examined — four studies, more than 72,000 participants, one of the largest controlled trials ever run plus two independent meta-analyses — the evidence points to the macro split being a non-factor for fat loss when overall food quality and calorie intake line up.

If keto works for you, the evidence says keep going. It produces equivalent fat-loss results. But it works because of the calorie deficit it creates, not because carbs were the enemy. If restricting carbs makes you miserable, the same evidence says a balanced approach delivers the same outcomes.

Pick the carb level you enjoy and can sustain. That is the evidence-based answer.

FitChef's meal plans set a minimum of 50 grams of carbs per day — not as a diet philosophy, but because fruit, vegetables, and legumes contain carbohydrates, and removing them doesn't improve fat-loss outcomes based on the evidence we examined. Among more than 40,000 members, 75% have weight loss as their primary goal. Most of them eat carbs.

One thing the diet-type debate completely misses: the structure of your food might matter far more than the macro split. In a controlled metabolic ward, people eating ultra-processed meals consumed over 500 extra calories per day compared to people eating whole food — with identical macronutrients, identical fiber, identical sugar.

Same carbs. Same fat. Same protein. But one group ate 500 calories more. The difference wasn't what was in the food. It was what the food did to how they ate.