Calorie deficit: the formula, the right size, and why the arithmetic drifts

A calorie deficit means taking in less energy than you use, so the body makes up the shortfall from stored fuel. Every diet that produces weight loss produces a deficit, whatever else it claims to be doing — keto, fasting, low fat, paleo, the one your colleague swears by. That is not a criticism of any of them. It is the reason to choose between them on the honest criterion, which is not which one is metabolically magic but which one you can actually live on.

This article is about the size of the deficit, the formula people use to set it, why that formula stops being true after a few weeks, and the three things that decide whether the weight you lose is fat or muscle.

The calorie deficit formula everybody quotes, and why it is wrong

"A pound of fat is 3,500 calories, so a 500 calorie daily deficit loses a pound a week." That figure comes from a calculation published by Max Wishnofsky in 1958, and as a first approximation it is fine. As a prediction over months it is not.

It fails for a straightforward reason: as you get lighter, a smaller body costs less to run. The same intake that produced a 500 calorie deficit at the start produces a smaller one at every subsequent weight, and the curve flattens. This is the arithmetic behind the plateau that people experience as a metabolic betrayal. Mostly it is not a betrayal. It is a smaller person.

Modern models — the NIH Body Weight Planner among them — predict a curve rather than a straight line, and the practical consequence is simple: expect the rate to slow, and plan a recalculation every ten pounds or so rather than treating the first number as permanent. The calorie calculator will give you the current one.

How big should a calorie deficit be

The defensible range is a loss of roughly 0.5 to 1% of body weight per week. For most people that is a deficit somewhere between 300 and 500 calories a day. For someone heavier it can be larger in absolute terms and still sit inside the same percentage.

Faster than that and three things happen together. A larger share of the loss comes from lean tissue rather than fat. Hunger rises steeply, because appetite regulation responds to the size of the deficit. And adherence collapses — which matters more than either of the others, because a diet you abandon in week three has a net effect of zero.

What about very low calorie diets

They have a legitimate place: under medical supervision, for specific clinical reasons, often before surgery or in the management of type 2 diabetes, where trials such as DiRECT have produced genuinely impressive remission rates. Those programmes involve monitoring, formula products designed to preserve protein and micronutrient intake, and a structured reintroduction of food.

Choosing one for yourself because it looks faster on a spreadsheet is a different proposition, and it is how the cycle described in the yo-yo effect gets started.

Three things that decide what you lose

Protein

Roughly 1.6 grams per kilogram of body weight, and higher — up to about 2.4 g/kg — for leaner people in a larger deficit. Protein does two jobs at once here: it signals the body to retain muscle rather than break it down for fuel, and it is the most satiating macronutrient per calorie, so it makes the deficit itself easier to hold. The protein calculator gives the range for your weight.

Resistance training

It burns very few calories, and that is not why it is on this list. It changes the composition of the loss. In trials comparing a deficit alone against a deficit plus resistance training, the training groups lose a similar amount of weight and considerably more of it as fat. Cardio is useful for the calorie side; lifting is what protects the tissue you want to keep.

Sleep

The most underrated item here. In a well-known controlled study, participants in the same deficit lost similar total weight on 8.5 hours and 5.5 hours of sleep — but the short-sleep group lost around 55% less fat and considerably more lean mass, while reporting more hunger. Sleep is not a lifestyle garnish on a diet; it is part of the mechanism.

Why calorie tracking fails, and what to do instead

Self-reported food intake is unreliable in a well-documented direction. Studies comparing food diaries against doubly labelled water — the gold standard for measuring energy expenditure — consistently find under-reporting, frequently in the range of 20 to 40%, and larger in people with higher body weight.

Neither dishonesty nor stupidity explains it. Portion estimation is genuinely hard, cooking oil and dressings are almost invisible in a diary, and the foods most often forgotten are exactly the calorie-dense ones. People are not lying about the salad. They are forgetting the two tablespoons of oil it was cooked in, which is 240 calories.

Two habits fix most of this without turning eating into accounting:

• Weigh food for two weeks, then stop. The point is not permanent tracking — it is recalibrating what 100 grams of pasta or a tablespoon of peanut butter actually looks like. Most of the benefit is in the recalibration, and it persists after the scale goes back in the drawer.
• Read the scale as a trend, not a verdict. Weigh at the same time daily and look at the weekly average. Day-to-day swings from water, glycogen, sodium and gut contents are routinely two to four pounds — large enough to hide a genuine fat loss for a fortnight and to trigger a pointless panic.

Menstrual cycles and the scale

Worth stating explicitly because it causes so much unnecessary discouragement: water retention across the luteal phase commonly adds one to four pounds that has nothing to do with fat. Comparing week to week at the same point in the cycle is the only fair comparison.

Adaptive thermogenesis: the part that is real and overstated

Beyond the arithmetic of a smaller body, energy expenditure does fall somewhat further than size alone predicts. This is adaptive thermogenesis, and it is real. The Biggest Loser follow-up study found substantial and persistent reductions in resting metabolic rate years after the competition, and that study is quoted constantly.

Two things are usually left out. It was an extreme intervention — enormous deficits and punishing exercise volumes over a short period — and other studies of moderate weight loss find much smaller adaptations. The other component is behavioural: non-exercise activity falls when people are in a deficit. They fidget less, walk less, take the lift. That is not a metabolic mystery, and it is partly addressable by deliberately keeping daily step count up.

The honest summary: metabolic adaptation is a headwind, not a wall, and it is a much bigger headwind after a crash diet than after a moderate one.

The part nobody sells

Maintenance is harder than the loss and it needs planning before the loss starts, not after it ends.

At the new lower weight, appetite hormones remain shifted toward eating more — leptin down, ghrelin up — and in some studies that shift persists for a year or more. Expenditure sits somewhat below what the new size alone would predict. Neither of those is a moral failing; both are physiology, and both are entirely manageable once you know they are coming.

Deciding in advance what the maintenance diet looks like — and treating it as the actual goal rather than the anticlimax after the interesting part — is the single clearest difference between a loss that holds and one that reverses. If you plan only the deficit, the deficit is all you get.