The nineties told you to cut fat at all costs. The keto movement told you fat was the answer and carbs were the real enemy. When the largest analysis of this question landed, covering 37 trials and over 57,000 people, it found both camps had been arguing about a variable that barely matters.
Cutting fat from your diet does produce weight loss. The nineties got that right. Across 37 trials and over 57,000 participants, reducing fat intake led to an average loss of about 1.4 kg, roughly three pounds. The weight of a bag of flour.
That's it. The entire result of the war on fat.
But here's what the nineties missed completely. The weight loss tracks the calories removed, not the fat itself. Each percentage point of calories shifted away from fat produces about 0.20 kg of weight change, and that lines up almost exactly with what basic calorie math predicts. Fat wasn't the villain. The calories packed inside it were.
The keto camp saw that clearly. In a metabolic ward study where every bite was measured, a 75% fat ketogenic diet didn't cause significant body fat gain. Fat isn't uniquely fattening. That much they got right.
Where keto went wrong was the next leap: claiming fat is metabolically special, that it unlocks some advantage other macros can't match. Thirty years of pendulum swings between "fat is evil" and "fat is magic."
The evidence says neither extreme was earned.
The Proof Comes From Every Direction
One study saying "it's about calories" could be a fluke. Five independent research teams saying it, using completely different methods, can't be.
An 811-person trial ran for two full years, comparing 20% fat diets to 40% fat diets. Both groups lost 3.3 kg. Not similar weight. Identical weight. A coin flip would produce less agreement than this result.
In a separate 12-month trial with 609 adults, led by Stanford researcher Christopher Gardner, neither the low-fat nor the low-carb group was told to count calories. Both spontaneously cut about 500 to 600 calories per day anyway.
The macro they restricted didn't matter. The calorie reduction was the active ingredient.
After a year, the difference between the two diets was 0.70 kg. Less than a jar of peanut butter. And here's what makes this finding sharper: the study was partially funded by NuSI, an organization that raised over $40 million to prove that carbs, not fat, are the problem. Their own funded research showed neither macro matters.
When 19 additional trials explicitly matched calories between groups, every comparison came up empty. No advantage for cutting carbs. No advantage for cutting fat.
And in controlled overfeeding, when researchers deliberately fed people more than they needed, the amount of body fat gained was the same regardless of whether the surplus came from fat, carbs, or protein.
Five lines of evidence. Five study designs. One answer.
The Insulin Theory Breaks Under Testing
If you've spent any time in keto or carnivore spaces online, you've absorbed a specific story: carbs spike insulin, insulin drives hunger, hunger drives overeating, and that's why carb-heavy diets make you fat. The carbohydrate-insulin model has billions of views on TikTok alone.
It makes intuitive sense. And it breaks every time researchers test it directly.
Kevin Hall's metabolic ward study, the most controlled dietary experiment in this evidence set, put participants on both a high-carb diet and a ketogenic diet for two weeks each. The high-carb side produced higher insulin, exactly as the model predicts.
But they ate 689 fewer calories per day.
Not on average across the group. Every single one of the 20 participants ate less on the high-carb diet. That's the equivalent of skipping an entire meal every day, without trying, despite the insulin the theory says should drive them to eat more.
A separate trial tested the hunger link head-on. 120 adults ate meals designed to produce different insulin responses. The insulin differences appeared on schedule. Hunger didn't change at all. The chain from insulin to hunger simply didn't hold.
In Gardner's year-long trial, researchers checked whether genetics or baseline insulin could predict who'd do better on which diet. Neither could. The model's predictions failed at every link in the chain.
insulin
What This Means for You
Fat doesn't make you fat. Calories do. But that sentence, true as it is, still leaves you with a question: what do I actually do with this?
If eating more fat helps you hit your calorie target because you feel fuller on fewer meals, the evidence supports that approach. If eating less fat helps because you can eat more food volume for fewer calories, the evidence supports that too. The macro you adjust matters far less than the calorie gap it creates.
And here's the number that reframes the entire debate. In one trial of 609 adults, the difference between the low-fat and low-carb groups was 0.70 kg after a year.
But individual outcomes on the exact same diet ranged from losing 30 kg to gaining 10 kg. A 40 kg spread. Individual variation was roughly 57 times larger than the diet-type effect.
You've been asking "which macro ratio is best?" The more powerful question is: what approach helps you stay in the right calorie range?
The pattern holds outside the lab, too. Macro customization data from over 40,000 FitChef members — 75% of whom are primarily focused on weight loss — shows the same thing the research predicts: how people split their fat and carb grams matters far less than whether they hit their calorie target.
But there's one question the weight evidence leaves on the table.
For weight on the scale, the type of fat you eat doesn't matter. A surplus calorie from olive oil and a surplus calorie from butter produce the same number.
But when researchers fed people identical calorie surpluses from saturated versus unsaturated fat and then scanned their bodies with MRI, the scale showed the same gain. How much was muscle versus how much was stored fat turned out to be dramatically different. What the scale shows and what the mirror shows aren't always the same thing.
That question deserves its own evidence deep dive.
Fat has 9 calories per gram versus 4 for carbs and protein. A tablespoon of olive oil packs roughly the same energy as a medium apple. When you add fat to a meal, you add calories in a small package. When you remove fat, you remove calories from a small package. The research found that the weight change followed the calorie change, not the fat change. In practical terms: the studies that swapped 2 tablespoons of oil for an extra cup of vegetables cut about 200 calories without changing the volume of food on the plate. The weight loss tracked the missing calories, not the missing fat.