What the evidence says about carbohydrates and physical performance — minus the noise.
For fifty years, one idea has sat at the center of sports nutrition: to perform, you need carbohydrates, and the harder you go, the more you need. Carb-loading, gels, the 120-grams-an-hour endurance protocol — all of it rests on that assumption. But how well does it actually hold up?
That was the question Bobby Patten and I took on in the first Ezz × Pegasus session — a two-hour conversation on carbohydrates and physical performance, recorded as preparation for our open-water swim in Sardinia. You can watch the whole thing here:
The argument, in brief
We came at the question from three directions.
Evolution. Long before agriculture, humans were endurance animals and fat-fueled hunters. The paleoanthropologist Miki Ben-Dor and colleagues, drawing on 25 separate lines of evidence, reconstructed where our ancestors sat in the food chain across the Pleistocene: our trophic level climbed steadily until, by the time of Homo erectus, we were hypercarnivores — apex predators drawing most of our energy from animal foods. We ran down large prey over hours, on foot, without weapons. Fat, not carbohydrate, was the ancestral fuel. Heavy carbohydrate consumption is a very recent chapter in the human story.
History. So where did "carb-loading" come from? Largely from a small set of studies in the 1960s — Swedish researchers using muscle biopsies on a handful of male graduate students — that showed high-carb diets let people exercise longer in the lab. Those findings were real, but narrow. What turned them into a cultural rule was the low-fat, high-carbohydrate revolution of the 1970s and 80s, scaled and amplified by an industry with an obvious interest in selling carbohydrate. The science was small; the marketing was enormous.
Modern evidence. More recent research keeps arriving at an inconvenient conclusion. Fat-adapted athletes perform well — in running (Prins and colleagues found no impairment in high-intensity efforts on a low-carb diet), in swimming (a randomized trial of elite open-water swimmers found carbohydrate feeding made no difference to speed or perceived effort), and in cycling (where the one measurable benefit was a tiny intra-race dose, roughly a couple of jelly beans an hour, in already-adapted athletes). Across the literature, habitual high-carb eating simply doesn't earn its reputation for performance. What it reliably does is raise blood sugar — sometimes, even in lean elite athletes, into pre-diabetic ranges.
The point isn't "carbs are bad"
It would be easy to hear all this as another diet crusade. It isn't. The honest conclusion — and the one the room, including a cardiologist who joined the conversation, kept returning to — is that nutrition is individual. It depends on your metabolism, your training, your goals, and your history. There is no single number that's right for everyone.
If there's a larger lesson, it's about how we know what we think we know. Nearly all sports-nutrition research is conducted on people — athletes, coaches, scientists — who have been carbohydrate-fueled their whole lives. It is genuinely hard to study a fat-adapted human when almost no one in the sample is one. So the science tends to confirm the world it was born in. The most important assumption in any field is often the one nobody thinks to measure, because everyone is standing in it.
That's the spirit of Ezz: maximum health, minimum noise. Not dogma in either direction — just a willingness to ask where the evidence actually points, and to be honest when the answer is "it depends."
Go deeper
This session was a preview of something we do in the field, not just on a screen. This November, the first Ezzpedition takes ten people into Big Bend, Texas — five days testing the functional capacity we spent this session talking about. There's a free online briefing in September for anyone curious.
- The first Ezzpedition — Big Bend, November: worldofezz.com/ezzpedition2026
- Test your health knowledge: worldofezz.com/knowledge
Reference: Ben-Dor M, Sirtoli R, Barkai R. The evolution of the human trophic level during the Pleistocene. American Journal of Physical Anthropology. 2021;175(S72):27–56.