BMI is your weight divided by your height squared, and that is the whole of it. It measures how heavy you are for your height — not how much fat you carry, not how fit you are, not how healthy you are. It was invented in the 1830s to describe the spread of body sizes across a population, and for that job it still works. Applied to one person it is a screening figure: a cheap first pass that says whether anyone should look closer, and nothing more.
How is BMI calculated?
In metric it is kilograms divided by metres squared. Someone 1.75 m tall weighing 70 kg has a BMI of 70 ÷ 1.75², which is 22.9.
In imperial the same ratio needs a constant: pounds divided by inches squared, multiplied by 703. The 703 is not a fudge factor or an adjustment for anything — it is the number that converts pounds per square inch into kilograms per square metre. Feed both formulas the same body and they agree to the decimal place anyone bothers reporting.
The arithmetic is trivial and the unit juggling is where people slip. A BMI calculator that accepts either system handles the conversion, tells you which band the result falls in, and works out what the healthy band weighs at your specific height. If you convert units by hand often, the metric-to-imperial conversions worth memorising covers the handful that keep coming up.
Where do the 18.5, 25 and 30 cutoffs come from?
The World Health Organization's adult bands are these:
| Category | BMI |
|---|---|
| Underweight | Under 18.5 |
| Healthy | 18.5 to 24.9 |
| Overweight | 25 to 29.9 |
| Obesity I | 30 to 34.9 |
| Obesity II | 35 to 39.9 |
| Obesity III | 40 and over |
WHO's own wording is drier than everyday usage: it calls the middle band the normal range and 25 to 29.9 pre-obesity. Either way the numbers are conventions. The relationship between BMI and health risk is a smooth curve with no steps in it, and the cutoffs are round numbers laid over that curve so that clinicians and public health agencies have something to count. Nothing about your body changes between 24.9 and 25.0. If you want to see what those bands weigh at your own height, the calculator lists all six in kilograms or pounds, which is a far more useful thing to look at than the label.
The ratio itself came from Adolphe Quetelet, a Belgian astronomer and statistician who was after a statistical description of the average man rather than a way to grade individual bodies. The name "body mass index" arrived in 1972, from the physiologist Ancel Keys, who compared several height-weight indices, found this one the best of a mediocre set, and said plainly that it suited population studies and not the evaluation of individuals. Neither of them designed it for the use it now gets.
What is BMI actually good for?
It is cheap. Two measurements, no equipment beyond a scale and something to stand against, no training required, and the result is comparable across millions of people measured by different hands in different countries. Nothing else that measures bodies is anywhere near as easy to collect.
And it does track risk at scale. The Global BMI Mortality Collaboration pooled individual data from 239 prospective studies covering 10.6 million adults and published it in The Lancet in 2016; all-cause mortality was lowest in the 20.0 to 25.0 range, and rose both above and below it. That finding is about groups. It tells you nothing about what will happen to any particular person in any particular band, and it is not a target. The headline analysis covered only never-smokers who had no chronic disease at recruitment and survived the first five years, which was 3.95 million of the 10.6 million. That filtering exists because illness and smoking drive weight down and death up at the same time, which otherwise makes the thin end of the curve look far more dangerous than it is.
What can BMI not see?
Four things, all of them consequential.
- What you are made of. The formula sees total mass. Muscle is denser than fat, so a trained athlete and a sedentary person of the same height and weight get the same BMI and are in completely different physical condition. Rugby forwards and heavyweight lifters routinely land in the obese band while carrying far less fat than that label implies.
- Where the weight sits. Fat around the abdomen carries more metabolic risk than the same mass on the hips or thighs, and BMI cannot distinguish them. This is the gap that waist measurement fills, and it is why NHS guidance in England now asks for both.
- Height, properly. Mass scales with volume, which grows with the cube of length, while BMI divides by a square. Take someone 1.60 m and 55 kg — BMI 21.5. Scale that body up to 1.90 m keeping every proportion identical and they weigh about 92 kg, for a BMI of 25.5. Same shape, different band. Real people are not scale models of each other so the effect is smaller than that in practice, but it runs in the same direction every time: tall people get pushed up, short people get pushed down.
- Who you are. At the same BMI, women carry more body fat than men, and a 70-year-old carries more fat and less muscle than a 25-year-old. An older adult can lose muscle, gain fat and hold an identical BMI throughout, which is the case where the number is not just imprecise but actively misleading.
Who do the standard thresholds not fit?
The 18.5 / 25 / 30 bands are for adults, and several common situations fall outside them:
- Children and teenagers. The expected ratio changes throughout growth, so they are assessed against BMI-for-age percentile charts. An adult BMI run on a twelve-year-old produces a number with no meaning attached.
- People of Asian descent. Body fat and cardiometabolic risk tend to be higher at a given BMI. A 2004 WHO expert consultation left the international cutoffs unchanged and added public-health action points at 23, 27.5, 32.5 and 37.5, for countries to set their own risk thresholds against.
- Pregnancy. BMI is uninterpretable during it. Clinicians use pre-pregnancy BMI.
- Amputation, significant fluid retention and muscle-wasting conditions. All of them break the assumption that weight stands in for tissue mass.
What should you measure alongside it?
Waist-to-height ratio is the cheapest thing you can add, and it catches exactly what BMI misses. The rule of thumb is a waist under half your height. NICE, which writes the NHS clinical guidance in England, classifies 0.4 to 0.49 as healthy central adiposity, 0.5 to 0.59 as increased and 0.6 or more as high, and asks for it alongside BMI in adults whose BMI is under 35. Above 35 it stops adding much, because the BMI has already said what there is to say.
Measure round the middle, midway between the bottom of your ribs and the top of your hip bone, at the end of a normal breath out. The weakness of this measurement is that tape placement varies a lot between people and between days, and it is easy to flatter yourself by half an inch without noticing. Take it the same way each time and watch the direction it moves rather than the absolute figure.
Beyond that you are into body composition measurement — skinfold calipers, bioimpedance, DEXA scans — and accuracy there costs either money or a trip somewhere. The cheap end is the inconsistent end, and the accurate end is not something you do at home on a Tuesday. For most people two crude numbers tracked over a year beat one precise number taken once.
So is your BMI worth knowing?
Yes, as one of two or three cheap numbers, and only if you read it as what it is. BMI answers the question "am I heavy for my height", which has a real answer. It does not answer "am I healthy", "how much fat am I carrying" or "should I be worried", and every argument about BMI comes from someone asking it the second set of questions and getting an answer anyway. A figure that arrives with no idea what your body is made of deserves to be treated as the start of a conversation with a doctor, not the end of one.
If you want the number rather than the arithmetic, the BMI calculator here takes metric or imperial, shows the WHO band, and lists what each band weighs at your height — nothing you type into it leaves your browser. It also prints the 23 and 27.5 action points next to the standard band. What it will not do is the second measurement: there is no waist field and no growth chart, so the ratio above is still you, a tape measure and a calculator, and a child's BMI still needs a percentile chart from a clinician.
If taking a formula apart to see what it actually claims is your kind of thing, how compound interest works does the same job on the other number people quote with more confidence than they have earned.
Frequently asked questions
What does BMI actually measure?
Weight relative to height, and nothing else — your weight in kilograms divided by your height in metres squared. It does not measure body fat, fitness or health, and it cannot tell muscle from fat. It was built as a way to describe populations, not as a verdict on an individual body.
Is a BMI of 26 bad?
It sits in the WHO overweight band, which starts at 25, but that boundary is a round number laid over a continuous curve. The difference in measured risk between 24 and 26 is small enough to vanish behind everything else about you: age, waist measurement, blood pressure, fitness. Read it as a prompt to check your waist-to-height ratio, not as a diagnosis.
Does BMI account for muscle?
No. The formula only sees total weight, and muscle is denser than fat, so a heavily trained person and a sedentary one of the same height and weight get an identical BMI. People in strength and contact sports regularly register as overweight or obese while carrying far less fat than those labels imply.
Is BMI accurate for tall people?
It runs slightly against them. Mass grows with volume, which scales with the cube of height, while BMI divides by height squared, so two people built to the same proportions get different numbers and the taller one gets the higher score. Real bodies are not scale models of each other, so the effect is smaller in practice than the geometry suggests, but it is consistent in direction.
Is waist-to-height ratio better than BMI?
It is better at the one thing BMI is blind to, which is where the weight sits. A waist under half your height is the simple version; NICE, which writes the NHS clinical guidance in England, treats 0.4 to 0.49 as healthy, 0.5 to 0.59 as increased and 0.6 or more as high, in adults with a BMI under 35. The catch is that tape placement is inconsistent, so measure it the same way each time and watch the trend.
Why is BMI still used if it is so flawed?
Because it costs nothing, needs two measurements anyone can take, and predicts risk well enough across large groups to be worth collecting. Pooled data from 239 prospective studies covering 10.6 million adults found all-cause mortality lowest between 20 and 25. That makes it a reasonable screening tool and a poor individual verdict, and most of the criticism of BMI is really criticism of the second use.
Last updated September 19, 2026