BMI: What It Is Good For, and What to Use Instead
BMI is not a bad measurement. It is a good measurement being asked the wrong question — it was built to describe populations, and it gets used to judge individuals.
In this guide
What BMI actually is
Body Mass Index is weight in kilograms divided by height in metres squared. It was devised in the 1830s by the Belgian statistician Adolphe Quetelet, who was studying the distribution of human physical characteristics across populations. He was not designing a clinical tool and said as much.
BMI = weight (kg) ÷ height (m)²
Under 18.5 underweight · 18.5–24.9 normal · 25–29.9 overweight · 30 and above obese. These cut-points are population-level statistical conveniences, not individual diagnoses.
Read that formula carefully and the fundamental problem is visible in it: there is no term for body composition. BMI cannot distinguish a kilogram of muscle from a kilogram of fat, and it has no idea where on your body that mass sits. Those two omissions are where every one of its failures comes from.
Where it fails
- Muscular people. A 180 cm athlete at 95 kg and 10 percent body fat has a BMI of 29.3 — “overweight”, one point from obese, while being leaner than nearly everyone around them. This is not an edge case; it applies to most trained men above average height.
- The “normal weight obese”. The inverse error, and the more dangerous one. Someone with low muscle mass and high body fat can sit comfortably inside the normal range while carrying a metabolically unhealthy amount of fat. BMI issues them a clean bill of health.
- Fat distribution is invisible to it. Two people at BMI 28 — one carrying it around the hips, one around the abdomen — have measurably different cardiometabolic risk. Visceral fat is the variable that matters, and BMI cannot see it.
- Height extremes. Because height is squared while human bodies do not scale quite that way, BMI tends to overstate adiposity in tall people and understate it in short people.
- Age and ethnicity. Older adults lose muscle and gain fat at stable weight, so an unchanged BMI hides a real change. And risk thresholds differ by ancestry — several health bodies apply a lower overweight cut-point of around 23 for South Asian populations.
Why it survives anyway
Given all that, it is worth asking why every clinic still uses it. The answer is that for its actual purpose it is excellent: it is free, needs no equipment beyond a scale and a tape, is perfectly reproducible, and correlates well with body fat across large groups. If you want to compare obesity prevalence between two countries or track it across a decade, BMI is the right instrument and always has been.
The failure is one of application. A measurement that is accurate about ten thousand people can be wrong about any one of them, and BMI is routinely used as though the first property implied the second.
Four better measurements
| Measurement | What it tells you | Cost | Reliability |
|---|---|---|---|
| Waist-to-height ratio | Central adiposity — the fat that carries the risk | A tape measure | High |
| Waist circumference | Visceral fat, tracked over time | A tape measure | High |
| Body fat percentage | Composition, which is what BMI is missing | Free (estimate) to a DEXA scan | Moderate to high |
| Progress photographs | Distribution and change your eye can actually judge | Free | High for change, poor for absolutes |
Notice what is absent: nothing here requires a laboratory. Three of the four need a tape measure and a mirror, which is the point — the reason BMI persists is convenience, and it turns out better measurements are nearly as convenient.
Waist-to-height ratio, the best single swap
If you replace BMI with exactly one number, make it this one. Measure your waist at the navel, divide by your height in the same units, and you have a figure that captures central adiposity — the fat deposited around the organs, which is the deposit most strongly associated with cardiometabolic risk.
| Ratio | Interpretation |
|---|---|
| Under 0.40 | Possibly underweight for your height |
| 0.40–0.49 | Healthy range |
| 0.50–0.59 | Increased risk — worth addressing |
| 0.60 and above | High risk |
The rule of thumb is memorable and hard to game: keep your waist under half your height. A 180 cm person wants a waist under 90 cm. A 165 cm person wants under 82.5 cm. Unlike BMI, this single number correctly classifies both the muscular athlete and the normal-weight-obese case, because it measures where the mass is rather than only how much of it there is.
How to use them together
No single number is enough, and the good news is that you do not need a single number. A five-minute monthly routine covers it:
- Weight, averaged weekly. Not for judgement — for detecting trends and recalculating your TDEE as it changes.
- Waist at the navel. Same time of day, relaxed, tape snug but not compressing. This is your primary health metric.
- Body fat estimate, from our body fat calculator using the US Navy circumference method. Treat it as a trend line rather than a truth — the estimate carries several percentage points of error, but its direction is reliable.
- Photographs, monthly, same light, same pose, same time of day. The most underrated measurement, because it is the only one that captures distribution.
We still show BMI in our TDEE calculator results, for one reason: people expect it and it provides rough context. But it sits beside the other numbers, not above them. If your BMI says overweight and your waist-to-height ratio is 0.45 and your lifts are climbing, the BMI is the number that is wrong.
Common questions
No — it is useful for the job it was built for. Across large populations it correlates well with body fat and tracks obesity trends reliably. It is unreliable as a verdict on any individual, because it cannot distinguish muscle from fat or see where fat sits.
Waist-to-height ratio, if you only want one number. Measure your waist at the navel and keep it under half your height. It captures central adiposity, which is the fat most associated with cardiometabolic risk, and it correctly classifies both muscular and low-muscle bodies.
Because BMI counts muscle as mass with no way to identify it. A muscular 180 cm person at 95 kg registers a BMI of 29.3 regardless of carrying 10 percent body fat. Check your waist-to-height ratio instead.
Yes, widely — it is free, reproducible, and a useful first screen. Many clinicians now pair it with waist circumference, and several health bodies apply lower thresholds for South Asian populations, precisely because BMI alone is too blunt.
Broadly, 10 to 20 percent for men and 18 to 28 percent for women is a commonly cited healthy range, with athletes often lower. Essential fat — below which health suffers — is roughly 3 percent for men and 12 percent for women.
Last reviewed September 14, 2026. Sources are cited inline; see our editorial policy for how we choose and review them.