Stronger by Math

Training decisions, run through the arithmetic

A note

The 3,500-Calorie Rule Is an Approximation, and It Fails in a Specific Direction

A pound of fat holds roughly 3,500 calories. Predicting weight loss from that number overshoots, and the reason is worth knowing.

By Tessa Lindqvist·Published ·293words

The 3,500-calorie rule is a reasonable constant used in an unreasonable way.

What it gets right

A pound of adipose tissue holds roughly 3,500 calories; a kilogram roughly 7,700. Over a short window that constant is a perfectly serviceable tool, and it is what the maintenance calculation on this desk is built on.

Where it fails

Naive extrapolation over six months

Deficit: −500 kcal/day → predicted 1 lb/week26 lb in 6 months

Actual, typically: 15–20 lb

Missing: expenditure fell with lost mass, spontaneous movement dropped, and intake was probably higher than logged

The constant is fine. The assumption that everything else stays still is what breaks, and it breaks progressively — which is why the prediction is good in month one and poor by month five.

The three leaks, sized

Expenditure falls with mass. A smaller body costs less to run. This is predictable and it means a fixed intake produces a shrinking deficit.

Spontaneous movement drops. The largest of the unpredictable effects, and mostly invisible to the person it is happening to.

The intake number is optimistic. This is usually the biggest of the three and the one nobody audits. If your tracker carries a 12% error band, on 2,000 calories that is ±240 — which can erase a modest deficit while the log shows perfect compliance.

How to use the rule properly

Over two to four weeks, to work backwards from your own data to your real maintenance figure. That is what it is good for, and it is genuinely useful there.

Not to predict six months of loss from day one. That prediction will be wrong, you will interpret being wrong as failure, and the constant was never the problem.

Questions we get asked

Is 3,500 calories really a pound of fat?

As a description of the energy in a pound of adipose tissue, roughly yes — the commonly used figure for a kilogram is about 7,700 calories. As a predictor of weight loss over months it overshoots, because it assumes your expenditure stays constant while you get smaller and move less, and neither holds.

Why am I losing less weight than the calculation says?

Three reasons, in rough order of size. Expenditure falls as you lose mass, so the deficit shrinks even at constant intake. Spontaneous movement drops during a diet. And the intake figure the calculation is built on usually carries more error than people assume — a 12% error band on 2,000 calories is ±240, comfortably enough to consume a modest deficit.

Does the 3,500 rule work at all?

It works well over short windows of two to four weeks, where the changes in expenditure are small. It is the basis of the maintenance-calculation arithmetic we use on this desk. It fails when extrapolated over six months, where its assumption of constant expenditure has quietly become false.

Tessa Lindqvist

Contributing writer, nutrition · Stronger by Math

Writes about intake, adherence and what the self-monitoring literature actually supports. Weighs her own food, which is why she is sceptical of estimates — including the good ones.

More from the desk