Four formulas, one height, four different answers. That is not a bug in the calculator — it is the honest state of the idea. There is no single ideal weight for a height, and the equations that claim to produce one were built for a purpose that has nothing to do with how you want to look.
At 175 cm, a man gets 70.5 kg from Devine, 68.9 kg from Robinson, 68.7 kg from Miller and 72.0 kg from Hamwi. They differ by 3.3 kg. The healthy BMI range at the same height is 56.7 kg to 76.3 kg — a band 19.6 kg wide. The formulas are arguing over the small number.
Where the four formulas come from
All four were published between 1964 and 1983, and all four are linear: a base weight at 5 feet, plus a fixed number of kilograms for every inch above it. They differ only in the base and the increment.
| Formula | Men at 5 ft | Per inch | Women at 5 ft | Per inch |
|---|---|---|---|---|
| Devine (1974) | 50.0 kg | 2.30 kg | 45.5 kg | 2.30 kg |
| Robinson (1983) | 52.0 kg | 1.90 kg | 49.0 kg | 1.70 kg |
| Miller (1983) | 56.2 kg | 1.41 kg | 53.1 kg | 1.36 kg |
| Hamwi (1964) | 48.0 kg | 2.70 kg | 45.5 kg | 2.70 kg |
The reason they agree as closely as they do is that they were fitted to the same thing: mid-century height-weight tables drawn from insurance data. A review of their origins concluded that the similarity between the equations came from the general agreement among those tables, and that any one of them may be used to estimate ideal body weight (Pai and Paloucek, Annals of Pharmacotherapy, 2000).
That review also explains what the equations were actually for. They entered clinical use because drug clearance correlated with ideal body weight, and the Devine equation in particular became the standard for pharmacokinetic dosing. So the number you are looking at was designed to help a pharmacist calculate a gentamicin dose — not to tell you what to weigh.
Why they disagree, and why the spread is the useful part
Each formula applies a different slope to the same height. Miller is the flattest — 1.41 kg per inch for men — so it gives the lowest answer at tall heights and the highest at short ones. Hamwi is the steepest at 2.70 kg per inch, so it diverges fastest as height rises. Devine sits in the middle and is the one most clinicians recognise.
None of this is precision. At 175 cm the four answers span 3.3 kg; at 230 cm they span 31 kg, because the slopes have been compounding for 40 inches. The formulas were never validated against measured body composition in a modern population, and they have no term for muscle, bone density, age, ethnicity or body fat.
So read the spread rather than the average. If four independent equations built on the same tables cannot agree within 3 kg on an average-height adult, no single "ideal weight" can be more precise than that.
A worked example
Devine: 50 + 2.30 × 8.9 = 70.5 kg
Robinson: 52 + 1.90 × 8.9 = 68.9 kg
Miller: 56.2 + 1.41 × 8.9 = 68.7 kg
Hamwi: 48 + 2.70 × 8.9 = 72.0 kg
Average: 70.0 kg. The four span 68.7 kg to 72.0 kg.
Healthy BMI band at 1.75 m: 18.5 × 1.75² = 56.7 kg, 24.9 × 1.75² = 76.3 kg.
Notice the asymmetry. The formula average, 70.0 kg, sits only 6.3 kg below the top of the healthy BMI band but 13.3 kg above the bottom. That is not a coincidence: the height-weight tables behind these equations described a population that was, on average, heavier than the bottom of today's healthy range. If you are a lean 60 kg at 175 cm, you are inside the healthy band and 10 kg below the "ideal" average — and nothing is wrong with you.
The BMI band is the better target
Body mass index is weight in kilograms divided by height in metres squared. The World Health Organization defines overweight as a BMI of 25 or above and obesity as 30 or above, which places the healthy adult band at 18.5 to 24.9.
For a given height, that band converts into a weight range:
| Height | BMI 18.5 | BMI 24.9 | Width of the band |
|---|---|---|---|
| 150 cm | 41.6 kg | 56.0 kg | 14.4 kg |
| 165 cm | 50.4 kg | 67.8 kg | 17.4 kg |
| 175 cm | 56.7 kg | 76.3 kg | 19.6 kg |
| 190 cm | 66.8 kg | 89.9 kg | 23.1 kg |
A 14 to 23 kg band is a much more defensible answer than a single figure. It also survives the things the formulas cannot handle: a 175 cm athlete at 80 kg with visible abs is outside the band, and a sedentary 175 cm adult at 74 kg with 30% body fat is inside it. BMI is a population screening tool, not a body composition measurement — it cannot tell those two people apart, and neither can any ideal weight equation.
Below 5 feet, the formulas stop working
Every one of these equations is anchored at 5 feet and adds weight for height above that point. Below 152 cm there is nothing to add, so all four return their 5-foot base weight regardless of how short you are.
You can see it in this calculator: a 150 cm man and a 120 cm man both get an average of 51.6 kg. At 120 cm that is a BMI of 35.8 — the tool's own BMI note says the healthy range at that height is 26.6 kg to 35.9 kg, which contradicts the formula output on the same screen.
If you are under 5 feet, ignore the formula results and use the BMI band. The equations simply were not built for you, and no amount of averaging fixes a missing term.
Frame size barely changes anything
The calculator lets you pick a small, medium or large frame, but it applies that adjustment only to Hamwi — plus or minus 10%. The other three formulas ignore it entirely, so switching frame moves one of the four numbers and leaves three unchanged.
This is roughly how the original Hamwi rule worked, and the ±10% convention is a rough proxy for skeletal build. It is also the weakest part of the whole exercise: wrist circumference or elbow breadth would be a measurement, whereas "small, medium or large" is a self-assessment, and most people pick the frame they want rather than the one they have.
What ideal weight is genuinely good for
It is not useless — it is just not a personal target. Ideal body weight is still used in clinical settings where a dose or a volume has to be scaled to body size rather than to actual weight, particularly for drugs that distribute poorly into fat tissue. In that context a stable, height-based figure is exactly what you want, and its imprecision is a known and accepted part of the calculation.
As a personal goal it fails for a simple reason: it answers "what did a mid-century insurance table consider average for your height", and you are asking "what should I weigh". Those are different questions.
What no formula can know
- Muscle mass. Two people at 175 cm can differ by 15 kg of lean tissue and both be healthy.
- Bone density and frame. A wrist measurement is a better proxy than a dropdown menu.
- Age. Weight at 25 and weight at 70 that are both healthy are not the same number.
- Ethnicity. WHO notes that BMI thresholds for overweight and obesity vary by population, and body composition differs at the same BMI.
- Pregnancy. Weight targets during pregnancy are set by a clinician from pre-pregnancy BMI, not by a formula.
- Fat distribution. Waist circumference and waist-to-height ratio say more about metabolic risk than weight alone.
If you want a number to work towards, use the BMI band as a wide target and your waist measurement as a second check. If you want to know what you are made of rather than what you weigh, body fat percentage and lean mass are the measurements that answer it.