7,700 kcal per kg of fat: where the number comes from

The figure traces to a single paper: Max Wishnofsky, "Caloric equivalents of gained or lost weight", American Journal of Clinical Nutrition, 1958, volume 6, pages 542–546. Wishnofsky concluded that one pound of body weight gained or lost is worth about 3,500 kcal. 7,700 is that number in metric — nothing was measured in kilograms, and no separate study produced it.

If you came here to cite something, cite Wishnofsky 1958. Everything else — the dietetic handbooks, the NHS-adjacent weight-management leaflets, the thousand calculator pages — is restating him. Below is how he got there, why the metric figure is 7,700 rather than 7,716 or 8,100, and what has been measured since.

How Wishnofsky arrived at 3,500

He did not run a trial. He reasoned from the literature available in 1958, in three steps:

• Human adipose tissue is roughly 87% fat — the rest is water, protein and connective structure. One pound is 454 g, so 454 × 0.87 ≈ 395 g of actual fat.
• Animal fat was taken to hold about 9.5 kcal per gram. 395 × 9.5 ≈ 3,750 kcal.
• He then applied conditions — adequate dietary protein, the subject in glycogen and nitrogen equilibrium — and settled on 3,500 as the working figure.

That last step is the one everybody forgets. Wishnofsky stated plainly that if his assumptions were not met, a 3,500 kcal deficit would not produce a pound of loss. The caveat was in the original paper. It did not survive the journey into popular use.

Why the metric figure is 7,700

One kilogram is 2.2046 pounds. 3,500 × 2.2046 = 7,716 kcal. Round that and you get 7,700 — which is why you also see 7,716, 7,720 and occasionally 7,000 in older material. They are all the same rule with different rounding.

You will also meet 7,200 kcal/kg. That comes from assuming a slightly lower lipid fraction or a 9.0 kcal/g energy density instead of 9.5, and it is no better or worse sourced than 7,700 — the inputs are estimates either way. Neither number is a physical constant, and quoting either to four significant figures implies a precision that does not exist.

Who repeats it, and what that means

The 7,700 kcal/kg equivalence appears throughout British and European dietetic teaching material, weight-management resources and nutrition coursework. That is genuine — it is the standard planning figure, and using it is not a mistake.

But a figure repeated by many organisations is not the same as a figure independently confirmed by many organisations. When you cite the British Dietetic Association or an NHS leaflet for 7,700, you are citing a restatement of Wishnofsky. If the citation matters — a dissertation, a paper, a clinical protocol — go to the 1958 source, and cite the modern critique alongside it.

"A number repeated everywhere for sixty years is still one number, from one paper, with its caveats stripped off."

What has actually been measured

Two pieces of work are worth having next to the original. Hall and Chow set out the mechanics in "Why is the 3500 kcal per pound weight loss rule wrong?" (International Journal of Obesity, 2013). Thomas and colleagues covered the clinical consequences in "Time to Correctly Predict the Amount of Weight Loss with Dieting" (Journal of the Academy of Nutrition and Dietetics, 2014).

The empirical number is the striking part. In the CALERIE calorie-restriction study, the measured energy content of weight change at week 4 was about 4,858 kcal per kg — well under two thirds of 7,700. Early weight loss is not pure fat: it carries water and glycogen, which are cheap. Later loss shifts toward fat and gets more expensive per kilogram, while total expenditure falls at the same time. The rule fails in both directions, just not simultaneously.

So 7,700 is a reasonable planning figure and a poor predictor. It is fine for setting a starting target. It is not fine for telling someone what the scale will say in eight weeks — which is exactly what most calculators built on it do.

How to cite it

For the original: Wishnofsky M. Caloric equivalents of gained or lost weight. Am J Clin Nutr. 1958;6(5):542–546. For the critique: Hall KD, Chow CC. Why is the 3500 kcal per pound weight loss rule wrong? Int J Obes. 2013;37(12):1614. Citing the pair is more defensible than citing either alone, because the honest position is "a widely used approximation with known limits", not "a constant".

How KusWise handles this

KusWise uses 7,700 kcal/kg for exactly one thing: turning a goal into a starting daily target. After that it stops trusting the arithmetic and starts trusting your data — it tracks the seven-day weight trend against what you actually logged and adjusts from the gap, rather than assuming the textbook conversion held. That is the same correction the research above argues for, applied weekly instead of never.

If you want the practical version of this, how many calories to lose 1 kg covers the deficit side, and our calculator will give you a Mifflin-St Jeor starting point in about thirty seconds. If you are unsure which expenditure number you are meant to be eating under, BMR vs TDEE settles that first.