DataFood
Educational only, and not medical advice. The figures on this site are estimates from population equations, not measurements of any individual. If you are pregnant, under 18, managing a medical condition, or have any history of disordered eating, speak to a clinician rather than a calculator.

About this site

What this site computes, where the equations come from, and what it refuses to do.

DataFood publishes nutrition arithmetic with its uncertainty attached. The calculations are the standard ones — Mifflin-St Jeor for resting expenditure, an activity multiplier for maintenance, grams of protein per kilogram of body weight — and the difference is that every figure arrives with the band it actually occupies.

Why that matters more than it sounds

A ten percent error on a maintenance estimate is about 230 calories for a typical adult. Someone eating at their estimated deficit who is at the low end of the band is losing weight quickly; someone at the high end is eating at maintenance and losing nothing, and will conclude after six weeks that their metabolism is broken. It is not broken. The estimate was a midpoint and nobody told them.

Where the equations come from

Mifflin-St Jeor is the equation clinical practice uses and the one with the smallest published error of the three in common circulation. The 7,700 calories per kilogram figure describes adipose tissue and is a simplification of real loss, which is a mixture of fat, some lean mass and a large amount of water in the first fortnight. The protein range of 1.6 to 2.2 grams per kilogram is where the evidence clusters for someone in a deficit who wants to keep muscle. Where a figure is contested, the page says so rather than picking the flattering end.

What it will not publish

No plan below 1,200 calories for women or 1,500 for men. No supplement, product or meal-delivery recommendation. No claim that a particular diet is superior, since the evidence on adherence is stronger than the evidence on macronutrient ratios. No before-and-after images and no rate of loss described as typical, because a typical rate is a statement about a population and every reader is an individual.

If food has stopped being simple

Calorie arithmetic is not help for disordered eating and can make it worse. Beat on 0808 801 0677 in the United Kingdom, and the National Eating Disorders Association helpline on 1-800-931-2237 in the United States, are free and confidential. They are listed here because a site of this kind that omitted them would be pretending it has no readers who need them.