How Statura's Calculators Work: Our Formulas and Sourcing

Most online calculators give you a number and nothing else. You cannot tell which formula produced it, whether that formula was appropriate for you, or how much confidence the result deserves.

We take the opposite approach. Every calculator on Statura names its formula, shows the equation, works through a numbered example with real values, and states plainly what the method cannot account for.

That transparency is not decoration. If you can see that your BMI came from dividing weight by height squared, you can also see why it says nothing about muscle — and that understanding is worth more than the number itself.

Nothing here is invented, adapted or 'improved'. Every calculation implements a published clinical or scientific formula exactly as its authors specified.

Several of these areas have more than one accepted equation, and they disagree. Our approach depends on the nature of the disagreement.

Where one formula is clearly better supported, we default to it and offer the alternatives. BMR is the clearest case: Mifflin-St Jeor is the default because it has the best validation record in the general population, with Harris-Benedict and Katch-McArdle available as alternates.

Where no formula is clearly superior — ideal weight, lean body mass — we show all of them side by side, plus their average. Hiding the disagreement would imply a precision that does not exist. Seeing a six-kilogram spread between four 'ideal' weights tells you something important about how much confidence any single one deserves.

No calculation on this site involves a server. When you type your weight into the BMI Calculator, the arithmetic happens in JavaScript on your own device, and the result never leaves it.

This is not incidental — it is why we can say plainly that we do not store your measurements. There is no account to create, no database recording your inputs, and no request carrying your height and weight anywhere.

It also means the calculators work instantly and continue working on a slow connection, which matters more than it sounds when the alternative is waiting on a round trip to check your BMI.

Every calculator works internally in metric, because the published formulas are specified in SI units. US customary inputs are converted to metric before any calculation, and results are converted back for display.

Rounding happens only at the final display step, never mid-calculation. Rounding intermediate values introduces errors that compound, and it is a common source of discrepancy between calculators that appear to use the same formula.

We also try not to imply more precision than exists. BMI is shown to one decimal place because the underlying measurement does not support two, and calorie targets are rounded because a figure like 1,847 kcal suggests an accuracy no prediction equation possesses.

Some omissions are as deliberate as the inclusions, and they are worth stating.

We do not offer pediatric BMI. Children and teenagers require BMI-for-age percentile curves from CDC or WHO growth charts, which is meaningfully more complex than adult BMI and easy to get subtly wrong. Rather than approximate it, we route under-20s to a note recommending a pediatrician.

We do not diagnose, and we avoid language that implies we do. A calculator can tell you that a measurement falls in a band associated with elevated population risk. It cannot tell you that you have a condition, and phrasing that blurs the line does real harm in health content.

We also do not gate anything behind an account, an email address or a paywall.

Every calculation function is covered by automated tests that run before any change ships. Those tests include reference cases with known correct answers — 175 cm and 70 kg must produce a BMI of 22.9; a 30-year-old man of 175 cm and 70 kg must produce a Mifflin-St Jeor BMR of about 1,649 kcal.

The tests also verify the boundaries: that invalid input produces a clear message rather than a nonsensical number, that unit conversion gives identical answers in metric and US customary, and that no calculator can display 'NaN' to a user.

This is ordinary engineering practice, but it is worth mentioning because a surprising number of health calculators fail exactly these cases.

Two commitments shape the interface as much as the formulas shape the numbers. The site targets WCAG 2.1 AA: every input has a real label, every result is announced to screen readers when it changes, every interactive element is reachable by keyboard with a visible focus ring, and no result is conveyed by colour alone — categories always pair a colour with an icon and a text label.

There is also an appearance panel offering theme, text size, contrast and motion preferences, because a site people read numbers off should adapt to the reader rather than the reverse.

On privacy, the calculators genuinely cannot collect your measurements, because there is nothing on the other end to collect them. Advertising is the only third-party code on the site, consent defaults to denied until you choose otherwise, and the privacy policy describes exactly what that involves.

Each calculator carries a 'reviewed for accuracy' date. Content is reviewed at least annually, and sooner when a source organisation updates its guidance — the IOM pregnancy ranges and WHO classifications are the most likely to change.

If you find an error, we would genuinely like to know. Health information that is wrong is worse than no health information, and a calculator that has been wrong for a year has been wrong for thousands of people.

Our full sourcing and review policy is on the Editorial Policy page, and every calculator links its formula back to the organisation or paper it came from.