Does counting calories work, or is it all about your hormones?

The science, in full

The debate has two named models, and they are worth stating fairly. The energy balance model holds that body weight is governed by the first law of thermodynamics: fat mass changes only when energy intake and expenditure diverge over time. It does not claim calories are consciously counted by the body. Rather, the brain integrates many internal and external signals, hunger hormones, palatability, food form and reward, to regulate how much we eat and how much we burn, with calories as the common currency of the outcome. The carbohydrate-insulin model, formalised by Ludwig and colleagues, reverses the usual arrow of cause. It proposes that refined, high-glycaemic carbohydrates raise insulin, insulin directs fuel into fat cells and suppresses its release, and the resulting shortage of circulating fuel drives hunger and lowers expenditure. In this view overeating is a consequence of fat storage, not its cause.

The strongest test of the mechanism is the metabolic ward, where intake is fully controlled. When calories and protein are matched and only the carbohydrate-to-fat ratio is changed, the large metabolic advantage predicted by a strong version of the carbohydrate-insulin model does not appear. In Hall and colleagues' inpatient study, switching men to an isocaloric ketogenic diet lowered insulin sharply but did not accelerate body-fat loss, and produced only a small, transient rise in expenditure of roughly a hundred calories a day that faded within weeks.

A meta-analysis of controlled isocaloric feeding studies by Hall and Guo pointed the same way, and if anything slightly favoured lower-fat over lower-carbohydrate diets for fat loss, the opposite of what the model predicts. Under conditions where calories are held constant, the insulin mechanism simply does not translate into meaningful differences in fat loss. This is the part of the argument where energy balance is on firm ground, and it is why claims of effortlessly burning more by cutting carbs alone should be treated with caution.

The carbohydrate-insulin model is not without support, and honesty requires saying where. In a weight-loss maintenance trial by Ebbeling and colleagues, participants on lower-carbohydrate diets appeared to have modestly higher total energy expenditure, on the order of a couple of hundred calories per day. That finding is genuinely contested: independent reanalyses questioned its magnitude and robustness, and it has not been cleanly replicated. This is the real frontier of the disagreement, and it is smaller and more technical than popular accounts suggest.

Where the two camps actually converge is appetite, and here diet composition clearly matters. Hall's inpatient trial of ultra-processed versus minimally processed diets matched calories, sugar, fat, fibre and macronutrients on offer, yet people spontaneously ate about five hundred calories more per day on the ultra-processed diet and gained weight. Food form and quality, not just macronutrient ratio, drove intake. In a later trial, a low-fat plant-based diet led people to eat fewer calories than an animal-based ketogenic diet, despite far more carbohydrate and higher insulin, which is difficult to square with a simple prediction that carbohydrate compels overeating.

Hormones also plainly regulate both sides of the balance, which is why the deficit is hard to sustain rather than hard to understand. Classic work by Leibel and Rosenbaum showed that losing weight lowers energy expenditure by more than body-composition change alone predicts, a phenomenon called adaptive thermogenesis. Sumithran and colleagues showed that after diet-induced weight loss, appetite hormones shift toward hunger, ghrelin rises while leptin and satiety signals fall, and these changes persist for at least a year. The body actively defends its former fat mass through hunger and energy thrift.

Modern pharmacology closes the loop and, tellingly, closes it through energy balance rather than around it. GLP-1 receptor agonists such as semaglutide produce weight loss of around fifteen percent in trials like STEP 1, and they do so principally by reducing appetite and food intake. This is a hormone acting on body weight in the clearest possible way, yet its lever is how much a person eats. It demonstrates both that hormones matter enormously and that they operate by moving the terms of the energy equation, chiefly intake.

So the framing of calories versus hormones is largely a false dichotomy. The energy balance is never violated, and there is no large hidden metabolic advantage from carbohydrate restriction at equal calories. But hormonal state and food quality powerfully shape hunger, spontaneous intake, satiety and the ability to maintain a deficit over time. The genuine scientific argument, well represented by Speakman and Hall on one side and Ludwig and colleagues on the other, is not whether thermodynamics holds but what controls the inputs to it, and how much specific weight to place on insulin and carbohydrate relative to the many other signals the brain reads.

For a person trying to lose weight, the practical reading is consistent with both models' better evidence. Calories are the accounting, and there is no way around them, but the useful leverage sits on the intake side: food quality, protein, fibre, sleep, and where appropriate, medication that lowers appetite. Individual variation is real and important, in the DIETFITS trial neither baseline insulin secretion nor a diet-matching genotype predicted who succeeded on low-fat versus low-carbohydrate eating, while adherence did. The best plan is the sustainable one, and if hunger or metabolic factors are working against you, that is a matter to discuss with your own doctor rather than to blame on willpower.

Putting this into practice. Instead of logging every bite, aim your energy at the food that keeps your hormones calm. Build each meal around protein, fibre and some healthy fat first, then add the carbohydrates, so your blood sugar rises gently rather than spiking and crashing two hours later. That steadier curve keeps hunger quieter, which makes eating a little less feel natural instead of a battle. Cut back on the fast sugars and refined carbs that drive the sharpest peaks. Counting calories now and then can be a useful eye-opener, but treat it as a tool for insight, not a lifelong chore. Let the quality of your food do most of the work.