Why does a calorie deficit stop working?

Where the numbers come from

Every calorie target starts from an estimate of what you spend, and it is worth knowing that it is an estimate rather than a measurement.

Total daily energy expenditure has four parts. Basal metabolic rate, the cost of simply being alive, is the largest at roughly 60 to 70 percent. The thermic effect of food — digesting and storing what you eat — is around 10 percent. Deliberate exercise is usually the smallest component, often under 10 percent for someone who is not an athlete. And non-exercise activity thermogenesis, or NEAT, covers everything else: walking, fidgeting, standing, gesturing. NEAT varies enormously between people, by several hundred calories a day, and it is the part that moves without you noticing.

The standard estimate of basal rate is the Mifflin-St Jeor equation, published in 1990, which takes weight, height, age and sex. It replaced the older Harris-Benedict equations because it was more accurate against the population it was validated on, but the honest description of both is that they predict an individual within about ten percent most of the time. On a 2,000-calorie maintenance figure that is a 200-calorie band, which is larger than many people set their deficit to be.

Total expenditure is then estimated by multiplying the basal rate by an activity factor, typically between 1.2 for sedentary and 1.9 for very heavy training. Those multipliers are coarse by construction. Choosing between 1.375 and 1.55 changes the answer by several hundred calories, and most people place themselves a tier too high, because activity feels more substantial than it measures.

The practical consequence is that a calculator gives you a starting hypothesis rather than a prescription. The correct use is to take the number, hold it for two or three weeks, weigh consistently, and adjust based on what actually happened. Your own trend line is a better measurement of your expenditure than any equation.

Why the arithmetic drifts

The 3,500-calories-per-pound figure traces to a 1958 paper by Max Wishnofsky, and it rests on a reasonable assumption: a pound of body fat stores roughly that much energy. As a description of the tissue it is broadly right. As a prediction of what a sustained deficit does over a year, it is not.

The reason is that the model is static and the body is not. A 500-calorie daily deficit predicts a pound a week indefinitely, and therefore 52 pounds in a year, which is far more than the same deficit reliably produces. Three things intervene.

The first is simple mass. A smaller body costs less to run, both at rest and in motion. As you lose weight your basal rate falls and every step burns fewer calories, so the deficit you set against your original expenditure shrinks against your current one until it closes entirely. This alone converts a straight line into a curve that flattens.

The second is adaptive thermogenesis: a reduction in energy expenditure beyond what the change in mass predicts. It is measurable, though the size is contested — studies commonly report somewhere between 50 and 150 calories a day beyond prediction during active weight loss, with larger figures after aggressive restriction. Part of it appears to be reduced NEAT. People in a deficit move less, spontaneously and without deciding to.

The third is that recorded intake and actual intake diverge. Self-reported food intake underestimates by a substantial margin in careful studies, frequently 20 percent or more, and the underestimate grows as intake grows. This is not dishonesty; it is portion estimation, cooking oil, drinks and the items eaten standing up. A deficit that stops producing results is more often a deficit that has quietly closed than a metabolism that has broken.

The useful reframing is that a plateau is information rather than failure. It means intake and expenditure have met, and the response is to remeasure both — recalculate against current weight, and tighten how intake is recorded for a fortnight — rather than to cut further on a target computed for a body you no longer have.

What the macro split is for

Splitting calories into protein, carbohydrate and fat does not change the energy total. It changes what the body does with a deficit, and one of the three has strong evidence behind it.

Protein matters most, for two reasons. In a deficit the body draws on both fat and lean tissue, and higher protein intake alongside resistance training measurably reduces the lean-tissue share of the loss. It also has the largest thermic effect of the three — roughly 20 to 30 percent of its energy is spent digesting it, against 5 to 10 for carbohydrate and 0 to 3 for fat — and it is the most satiating per calorie. Common recommendations for someone losing weight while training sit between 1.6 and 2.2 grams per kilogram of body weight per day, well above the 0.8 figure that represents a minimum to avoid deficiency rather than an optimum.

Fat has a floor rather than a target. It carries the fat-soluble vitamins and is required for hormone synthesis, and intakes below roughly 20 percent of calories are where problems begin to appear. Beyond that floor, the split between remaining fat and carbohydrate is largely a question of what keeps you full and what fuels your training, and controlled studies that match calories and protein find little difference in fat loss between higher-carbohydrate and higher-fat arrangements.

Fibre deserves a mention that macro calculators rarely give it. Around 25 to 30 grams a day supports satiety and digestion, and it is the single easiest thing to lose when calories are cut, because the foods that carry it are bulky and the foods that fit a tight target often are not.

Calories, kilocalories and kilojoules

The unit confusion here is genuine and worth clearing up, because it appears on every food label.

What everyone calls a calorie in a nutrition context is properly a kilocalorie — the energy to raise a kilogram of water by one degree Celsius. The lowercase calorie is a thousandth of that. Nutrition labels and calculators use the kilocalorie throughout and simply write it as Calorie or kcal, so a 2,000-calorie diet is two million of the physics unit. Nothing in practice depends on the distinction, but it explains why the two spellings appear side by side.

The kilojoule is the SI unit and the legal one in Australia, New Zealand and the European Union, where labels carry both. The conversion is fixed: one kilocalorie is 4.184 kilojoules. So 2,000 kcal is 8,368 kJ, and an 8,700 kJ figure — the reference intake printed on Australian packaging — is about 2,080 kcal. Multiply by 4.184 going to kilojoules, divide going back.

Hydration targets deserve the same scepticism as calorie targets. The eight-glasses figure has no identifiable source in the research literature. Institutional guidance is expressed as total water from all sources, including food, and sits near 3.7 litres a day for adult men and 2.7 for adult women — of which roughly a fifth arrives in food. Requirements move substantially with body size, climate and activity, which is why a calculator that takes weight and activity gives a more honest answer than a fixed number of glasses.

Questions