Ideal Weight Calculator
Compare ideal body weight across the Devine, Robinson, Miller and Hamwi equations, plus a healthy BMI range for your height. Seeing all four at once shows how much these clinical estimates actually disagree.
Compare four ideal weight formulas
Enter sex and height. Current weight is optional. Add it to see the gap to target.
Average of the four
A mean of four disagreeing equations is a convenience, not a consensus.
Devine 1974
Written for gentamicin dosing. Still the hospital default.
Robinson 1983
A refit of Devine against the 1983 Metropolitan Life tables.
Miller 1983
Highest starting weight, shallowest slope. It favours short frames.
Hamwi 1964
The oldest of the four, and the one dietitians still teach.
How ideal body weight is calculated
All four equations share a single shape. Each fixes a base weight at a height of five feet, then adds a fixed number of kilograms for every inch above that. Nothing else enters the calculation: not your age, not your build, not your ancestry, not what you weigh today. Two people of the same sex and height always receive the same answer.
Devine (1974) men: 50.0 + 2.30x · women: 45.5 + 2.30x
Robinson (1983) men: 52.0 + 1.90x · women: 49.0 + 1.70x
Miller (1983) men: 56.2 + 1.41x · women: 53.1 + 1.36x
Hamwi (1964) men: 48.0 + 2.70x · women: 45.5 + 2.20x
Example for a man of 175 cm (68.9 in, so x = 8.9):
Devine 50.0 + 2.30 × 8.9 = 70.5 kg
Robinson 52.0 + 1.90 × 8.9 = 68.9 kg
Miller 56.2 + 1.41 × 8.9 = 68.7 kg
Hamwi 48.0 + 2.70 × 8.9 = 72.0 kg
The calculator converts imperial entries to centimetres first and then back to inches, so both unit systems run through identical arithmetic and return identical results. Weights are shown in pounds when you switch to imperial, but the equation itself is always solved in kilograms.
Sources: Devine BJ, Drug Intelligence and Clinical Pharmacy 8:650–655 (1974); Robinson JD et al., American Journal of Hospital Pharmacy 40:1016–1019 (1983); Miller DR et al., American Journal of Hospital Pharmacy 40:1622 (1983); Hamwi GJ, in Danowski TS (ed.), Diabetes Mellitus: Diagnosis and Treatment, American Diabetes Association (1964). Historical account: Pai MP & Paloucek FP, Annals of Pharmacotherapy 34:1066–1069 (2000).
Why four formulas give four answers
The four equations were fitted independently, to different reference data, for slightly different purposes. None was ever validated against a health outcome. Laid out together, the pattern in their coefficients explains almost everything about how they behave.
| Equation | Base at 5 ft (men) | Per inch (men) | Behaviour |
|---|---|---|---|
| Hamwi (1964) | 48.0 kg | +2.70 kg | Lowest base, steepest climb; most generous to tall people |
| Devine (1974) | 50.0 kg | +2.30 kg | The middle of the pack, and the clinical default |
| Robinson (1983) | 52.0 kg | +1.90 kg | Deliberately flatter than Devine at the tall end |
| Miller (1983) | 56.2 kg | +1.41 kg | Highest base, flattest climb; most generous to short people |
Because the bases and the slopes run in opposite directions, the rankings invert as you move up the height scale. For a man of 165 cm, Miller returns the highest figure of the four and the whole set falls within about 1.8 kg. By 175 cm, Miller has become the lowest. At 190 cm the four answers span 77.1 kg to 88.0 kg, a disagreement of nearly 11 kg about the same body. There is no published basis for choosing between them, which is why this page shows all four rather than one.
They were written to calculate drug doses
This is the fact that reframes everything else on the page. Devine's equation first appeared in 1974 in Drug Intelligence and Clinical Pharmacy, in a discussion of how to dose gentamicin, an aminoglycoside antibiotic with a narrow therapeutic window and real kidney toxicity if you overshoot. Gentamicin distributes poorly into fat, so dosing an obese patient on total body weight overdoses them. Devine needed a quick, reproducible estimate of the lean-ish mass the drug would actually reach, and proposed one. As Pai and Paloucek documented in their 2000 review of the equation's origins, it was offered without a derivation, without a validation cohort, and without any claim that it described health.
Robinson and Miller both published in 1983 in the same journal, in the same field, for the same reason: pharmacists wanted a better fit to the Metropolitan Life insurance tables then in use. Hamwi's version, older still, came out of a 1964 American Diabetes Association chapter on diet therapy, as a bedside shortcut a clinician could do in their head while writing a meal plan.
So the lineage is pharmacy and dietetics, not physiology. That origin explains the omissions people find surprising. The equations ignore frame size because a drug dose does not need it. They ignore muscle mass, ethnicity and age for the same reason. They are step functions in inches because inches were what a chart on a hospital wall gave you. Read as what they are, a dosing heuristic, they work perfectly well. Read as a personal goal weight, they are being asked a question their authors never posed.
Where these numbers mislead
Every limitation below is a direct consequence of an equation that knows only your sex and your height.
- They cannot see body composition. A 180 cm man carrying 82 kg of well-trained muscle and a 180 cm man carrying 82 kg with very little muscle receive an identical verdict. Since the whole reason weight matters to health is what the weight is made of, this is the deepest problem with the concept.
- Below 5 ft they stop working. Each formula is anchored at 60 inches and only adds from there; none of the original papers defined behaviour underneath it. This calculator holds the base constant below 152.4 cm, which is the standard clinical convention, but the figure it returns is an extrapolation rather than a result. Some hospital protocols instead subtract 2 to 5 kg per inch below 5 ft, and those protocols disagree with each other too.
- They disagree most where you need them least. At exactly 5 ft, before height has contributed a single kilogram, the four equations already give a woman 45.5 kg, 49.0 kg, 53.1 kg and 45.5 kg. That 7.6 kg gap is pure difference of opinion between the authors.
- No frame-size adjustment survives. Hamwi's original method was not a single number: it came with an instruction to adjust the result by roughly 10% up or down for a large or small skeletal frame, judged from elbow breadth or wrist circumference. Almost every modern implementation, including this one, drops that step, which quietly removes a 14 kg band of legitimate variation from a 70 kg answer.
- No ethnic variation. The reference data behind all four sets of coefficients came from mid-twentieth-century North American insurance populations. Body proportions, limb-to-trunk ratios and the body fat carried at a given weight all vary between populations, and none of that is represented.
- They are silent on age, pregnancy and amputation. Lean mass falls steadily after roughly age 50, so a fixed target set at 30 is not the right target at 70. The equations have no meaning during pregnancy, and clinical practice applies separate correction factors after limb loss.
A better target than a single number
The honest replacement for an ideal weight is a healthy weight range, and the calculator shows one at the top of the results: the span of weights that would put your height inside the WHO healthy band of BMI 18.5 to 24.9. For a 175 cm adult that runs from about 56.7 kg to 76.3 kg. It is a wide band, and the width is the point: it is the honest expression of how much healthy variation exists between two people of the same height.
Notice how the two approaches relate. Every one of the four equation results for a 175 cm man falls inside that BMI band, near its upper-middle. The equations are not wrong so much as needlessly precise: they pick one point out of a legitimate span of nearly 20 kg and print it to a decimal place. A number that specific implies a level of knowledge about your body that no formula using only your height can possibly have.
The second measurement worth having costs nothing and adds what BMI and ideal-weight equations both miss. Wrap a tape around your waist at the midpoint between the lowest rib and the top of the hip bone, then divide that measurement by your height. A waist-to-height ratio below 0.5 (keeping your waist under half your height) is the threshold used in UK NICE guidance and supported by Ashwell and Gibson's 2016 analysis in BMJ Open, which found the ratio outperformed BMI as a predictor of cardiometabolic risk. It works in any unit system and needs no chart.
In practice, three things beat a target weight. Track the direction your weight is moving over months rather than its distance from a formula. Measure your waist every few weeks, because a falling waist with a stable weight is the signal you actually want. And check the things weight is only a proxy for: blood pressure, fasting glucose, a lipid panel. Our BMI calculator gives you the range shown above with the Asian-Pacific thresholds alongside it, and the body fat calculator takes the tape-measure step that separates a heavy body from a fat one.