What BMI cannot tell you
BMI is weight divided by height squared. That is the entire measure — one number in, one number out, no account of muscle, frame, age, sex or where the weight sits.
It survives because it is cheap and it works reasonably well across large groups. Applied to one person it can be badly wrong in both directions, and there is a measure that is just as easy and rather better.
What the number is
Someone 175cm tall weighing 70kg has a BMI of 22.9, which sits in the healthy band of 18.5 to 25. Read the other way, that height has a healthy range of about 56.7 to 76.3kg — a spread of nearly twenty kilograms, which is worth noticing before treating any single reading as precise.
The measure was devised in the nineteenth century for describing populations, not diagnosing individuals, and it was never intended to do the job it is now used for. The bands themselves are drawn from mortality data across large groups, which is exactly the setting where an average-shaped assumption is safe.
For an individual it is a screening tool. A reading well outside the bands is worth following up; a reading inside them is not a clean bill of health.
Where it goes wrong
Muscle is denser than fat, so it goes wrong most obviously on muscular people. A well-trained athlete regularly reads as overweight while carrying very little fat, because the formula has no way to know what the weight is made of.
It goes wrong the other way too, and this direction gets far less attention. Someone sedentary with little muscle can sit comfortably inside the healthy band while carrying a proportion of body fat that would concern anyone measuring it directly. That is a false reassurance, and it is the more common error.
The third problem is that it cannot see distribution. Two people with identical BMIs can carry their weight in completely different places, and abdominal fat is the part that carries the metabolic risk. The formula treats those two people as the same.
For children none of this applies at all — they are assessed against age and sex percentile charts, which is a clinical judgement rather than an arithmetic one.
Waist to height: simpler, and better
Keep your waist under half your height. That is the whole rule. It needs no calculator, works in any unit because it is a ratio, and tracks health outcomes better than BMI does.
Someone 180cm tall wants a waist under 90cm. At 85cm the ratio is 0.47, comfortably inside. The reason it outperforms BMI is precisely the thing BMI cannot see: it measures where the weight is rather than how much there is, and abdominal fat is what the risk attaches to.
The measurement itself is the part people get wrong. Measure at the midpoint between the lowest rib and the top of the hip bone, at the end of a normal breath out. Not at the trouser line, which sits lower and reads smaller, and not held in.
Height 180 cm -> half is 90 cm
Waist 85 cm -> ratio 0.47
Under 0.5 is the target, in any unit
Work it out: Waist to Height Ratio Calculator →
Body fat estimates are estimates
Tape-measure methods — the Navy formula and its relatives — work from neck, waist and hip circumferences and are considerably more informative than BMI, because they are at least trying to measure composition.
They are still estimates fitted to population data, and they carry a few percentage points of error on any individual. That is fine for the job most people want: tracking a direction over months. It is not fine for treating a single reading as a fact about your body.
The practical advice is to measure the same way, at the same time of day, with the same tape, and to care about the trend rather than the number. A method with consistent bias is perfectly good at showing change even when it is poor at showing level.
Questions
What is a healthy BMI?
18.5 to 25 for adults. At 175cm that is roughly 56.7 to 76.3kg — a range of nearly twenty kilograms, which is a reminder of how coarse the measure is.
Why does BMI say I am overweight when I am not?
Because it cannot tell muscle from fat. Weight divided by height squared is all it has, so a muscular person reads high while carrying very little fat.
Can BMI be too reassuring?
Yes, and this is the more common failure. Someone with little muscle can sit inside the healthy band while carrying a concerning amount of body fat, because BMI has no way to see composition.
What is better than BMI?
Waist-to-height ratio: keep your waist under half your height. It needs no calculator, works in any unit, and tracks outcomes better because it measures where the weight sits rather than only how much there is.
Where exactly do I measure my waist?
At the midpoint between the lowest rib and the top of the hip bone, at the end of a normal breath out. Not at the trouser line, which sits lower and flatters the reading.
How accurate are tape-measure body fat estimates?
Within a few percentage points, which is good enough to track a trend and not good enough to treat a single reading as fact. Measure consistently and watch the direction.
Does BMI work for children?
Not in the same way. Children are assessed against age and sex percentile charts, which is a clinical judgement rather than a calculation.