Three calculators. Each one shows the formula it used and names the source that formula comes from, runs entirely on your device, and asks you for nothing. None of them hands you a target you did not set.

What is not here, and why

There is no BMI calculator on this site and no body fat calculator. Both are decisions, not gaps, and they are the two biggest search terms in this whole cluster — so skipping them is not free and we would rather say so than leave a hole.

A BMI calculator’s entire output is a verdict about a body. This product is designed against a systematic review of 38 studies in Body Image which found that regular calorie-app use correlates with higher disordered-eating symptoms, worst in adolescents. When we harvested what people actually type around these calculators, two of the ten live suggestions under bmi calculator were bmi calculator children and bmi calculator teen. That is precisely the audience the finding is about.

Body fat is the same objection with an extra problem on top: every method a web calculator can use needs tape measurements that most people estimate, so the number carries an error nobody can see.

If you want either number, plenty of sites will give it to you. What we will not do is build the one thing this app exists to argue against and then ask you to trust the app.

What the three that are here cannot tell you

Your number, specifically. A formula is a population average with your figures substituted in. Two people the same height, weight, age and sex can differ by several hundred calories a day, and no equation on this site can say which of them you are.
Anything about anyone under 18. Mifflin-St Jeor was fitted on adults. The pages say so, and they mean it.
Anything medical. None of this is medical advice, and none of it is built for anyone treating an eating disorder.
What you actually ate. Which is the harder half of the problem, and the one the app is for.

Why a calorie app is publishing calculators

Honestly: because they are the front door. Somebody working out a protein target has already decided the question matters to them, and the next thing they hit is the part no formula answers — how much protein was in that. We would rather meet that person with a straight answer and then say what we build, than intercept them with a landing page.

Demand measured 12 September 2026: 222 Google autocomplete suggestions across 23 seeds in en-us, zh-TW, ko-kr and ja-jp, committed to the repository as `research/SNAPSHOT-CALC-2026-09-12.jsonl`. Autocomplete shows phrasing and breadth, not volume, and no volume figure appears on any of these pages. Disordered-eating finding: Flinders University systematic review, 38 studies, Body Image.