What this page gives you, and why it is two answers rather than one
1. Enter your height
Centimetres, or feet and inches. Height is doing almost all of the work here — three of the four classic formulas use nothing else except your sex — so measure without shoes and be honest about the half-inch.
2. Choose your sex
All four formulas are sex-specific, with a different base weight for men and women at five feet. This is a limitation of the formulas as published in the 1960s to 1980s, not a claim by us about how bodies work, and it is one of several reasons to treat their output as a rough landmark.
3. Enter your age, because it decides which standard applies
Not decoration, and not used by any of the four formulas — none of them contains an age term. It is used by the parts of the page that can honestly act on it. Below 20 the adult standards are withdrawn entirely rather than printed with a caveat. At 65 and over, an extra reference point appears, because the evidence about where the healthy range sits genuinely changes in later life.
4. Add your current weight if you want the comparison
Optional. Enter it and the tool shows where you actually sit relative to both answers, plus your BMI — which the previous version never displayed even though it had every number it needed.
5. Measure your wrist for a frame-size adjustment
Also optional, and more useful than it sounds. Wrist circumference is mostly bone and skin, so it is a reasonable proxy for skeletal frame. Hamwi’s original method came with a plus-or-minus ten per cent adjustment for frame size, and the wrist measurement is how that adjustment was meant to be made.
6. Read the healthy BMI range first
It is the answer that is defined at every height, derived from the classification actually used in clinical practice today, and expressed as a range rather than a single figure. If you only take one number away from this page, take this one.
7. Then read the four formulas, and their spread
Devine, Robinson, Miller and Hamwi are shown separately with the gap between the highest and lowest called out. That spread is the honest headline: four published formulas for the same quantity, disagreeing by several kilograms at ordinary heights.
Devine was written to calculate drug doses, not health targets
This is the single most useful thing to know about where “ideal body weight” numbers come from, and the previous version of this page never mentioned it.
In 1974, the pharmacist Ben Devine published a formula in a short article about gentamicin dosing. Gentamicin is an antibiotic distributed mainly through lean tissue rather than fat, so dosing it on total body weight overdoses heavier patients — a real danger with a drug that can damage hearing and kidneys. Devine needed a quick bedside estimate of lean mass from height. He produced one: 50 kg plus 2.3 kg for every inch over five feet for men, 45.5 kg plus 2.3 kg per inch for women.
It was never validated as a health target, never derived from outcome data, and was not intended to tell anyone what they should weigh. It spread because it is easy to do in your head, and it remains genuinely standard in clinical pharmacy for exactly the purpose it was designed for. Its second career as an “ideal weight” is an accident of convenience.
Robinson (1983) and Miller (1983) are both revisions of Devine fitted to different reference data. Hamwi (1964) predates all of them and came from diabetes practice, where it was used to estimate calorie prescriptions — and it was always meant to be used with a frame-size adjustment of roughly ten per cent in either direction, which most calculators quietly drop.
Why every height below five feet used to give the same answer
All four formulas share a shape: a base weight at exactly five feet, plus a fixed increment for each inch above that. None of them defines anything below five feet, because none of the reference populations they were fitted to extended there in useful numbers.
The previous build handled this with over = Math.max(0, inches − 60). Clamping at zero means that below five feet the increment term vanishes and only the base weight remains — so a woman of 135 cm and a woman of 152 cm were both told their ideal weight was 45.5 kg. Seventeen centimetres apart, identical answer, no warning of any kind. For a site with a large South Asian readership, where adult heights below five feet are common, that was not a rare edge case.
The tool now says plainly that the formulas are undefined below five feet and shows the healthy BMI range instead, which is defined at any height because it is a ratio rather than a base-plus-increment rule. That is the correct answer rather than a padded one.
The old headline was a union of extremes wearing the word “combined”
The previous version computed the lowest of the four formulas and the BMI floor, then the highest of the four and the BMI ceiling, and described the resulting span as “combining four classic formulas with the healthy-BMI range”.
Taking the minimum and maximum across a set of sources is not combining them. It cannot narrow an estimate; adding a fifth source can only ever leave the range the same or make it wider. And when we measured what it actually did across twenty-four height and sex combinations, the four formulas changed the displayed range in three of them — and in those three only because of the sub-five-foot clamping bug described above. For every ordinary adult height, the visitor was shown the plain healthy-BMI band and told it was the consensus of five different methods.
The two answers are now reported as two answers. They measure different things and they were built for different purposes, so making them agree would require hiding something.
Reading the healthy weight range from BMI
Because BMI is weight divided by height in metres squared, any height has a corresponding weight range for the healthy band of 18.5 to 24.9. At 165 cm that is 50.4 kg to 67.8 kg. At 5 ft 10 in it is 128.9 lb to 173.5 lb.
That range is wide on purpose. A span of seventeen kilograms at average adult height is the honest width of “healthy” as the evidence actually supports it, and it is a useful corrective to the whole idea of an ideal weight. There is no single correct number. There is a band, and within that band your build, your muscle mass and where you carry fat matter more than the exact figure.
If you are of South Asian, East Asian or South-East Asian descent, the tool can apply the WHO 2004 action point of 23.0 as the upper bound instead of 24.9. That recommendation came from a WHO expert consultation published in The Lancet in 2004, which found that cardiometabolic risk begins to rise at a lower BMI in many Asian populations. It narrows the range, and the tool tells you when it has applied it.
What your wrist has to do with it
Frame size is the missing variable in every height-based formula. Two people of exactly the same height can have visibly different skeletal builds, and a target weight that ignores this will be too low for one and too high for the other.
Wrist circumference is the traditional proxy, because there is very little muscle or fat there to confuse the measurement. The Grant method compares your height divided by your wrist circumference against reference values to classify your frame as small, medium or large. Hamwi’s formula was published with a plus-or-minus ten per cent frame adjustment precisely because the author knew a single figure per height was insufficient; most online calculators reproduce his numbers and drop his caveat.
Measure just past the wrist bone, on the hand side, snug but not tight. It is optional here, and if you skip it nothing about frame size is claimed.
Where an “ideal weight” is the wrong question
If you carry a lot of muscle, every formula on this page will read low, because they were all fitted to general populations rather than trained ones. A weight several kilograms above the formula figure may represent better health than the figure itself.
If you are older, the evidence does not support driving toward the bottom of the healthy band. Muscle and bone are lost with age, and in later life a slightly higher weight is associated with better outcomes, not worse ones.
If you are pregnant, none of this applies. Weight gain in pregnancy is assessed against pre-pregnancy BMI on an entirely separate framework — our pregnancy weight gain calculator uses it.
If you are under 20, adult targets do not apply at all; growth charts and percentiles do.
If you have a history of disordered eating, a page that produces a target number can be actively unhelpful. There is no shame in closing it. A doctor or a registered dietitian is the right place for this conversation, and the numbers here are not a goal anyone has set for you.
Why age changes the answer, and why the formulas cannot see it
Age is a genuine factor in what a healthy weight looks like, and it is worth being precise about where it enters and where it does not.
It does not enter the four formulas. Devine, Robinson, Miller and Hamwi contain no age term whatsoever. A muscular twenty-five-year-old and a sedentary seventy-five-year-old of the same height and sex receive an identical number from all four. That is a limitation of the formulas as published, not something a calculator can quietly correct, and pretending otherwise would mean inventing a standard that does not exist.
It does not enter the WHO adult classification either, and this surprises people. The band of 18.5 to 24.9 is deliberately not age-stratified; it applies unchanged from twenty upwards. The 1990 edition of the Dietary Guidelines for Americans did once publish age-adjusted ranges — a BMI of 19 to 25 for ages 19 to 34, and 21 to 27 for 35 and over — but that distinction was withdrawn in the 1995 revision, and the NHLBI adopted a single adult range in 1998. Any site still presenting age-banded BMI targets as current is quoting a standard that was retired thirty years ago.
Where age genuinely does change things is at the two ends, and that is exactly where this tool now acts on it.
Under twenty, the adult standards are withdrawn rather than caveated
Enter an age below twenty and this page stops giving adult answers. The formula figures go, the adult healthy band goes, the chart goes, and an explanation takes their place.
This is not excessive caution. Body composition changes continuously through growth, so a fixed weight range is meaningless for someone still growing. Young people are assessed instead on BMI-for-age percentiles, which compare an individual against others of the same age and sex — the CDC growth charts, published in 2000 and still the United States reference, and the WHO growth reference for children and adolescents aged 5 to 19, released in 2007. On those charts the healthy band is the 5th to the 85th percentile, not a figure in kilograms, and it shifts month by month as a young person grows.
A calculator that prints an adult ideal weight for a fifteen-year-old is producing a confident wrong answer, and a confident wrong answer aimed at a teenager about their body weight is worse than no answer at all. A doctor, school nurse or paediatrician can plot a growth chart properly, and that is the right place for this question.
At sixty-five and over, the risk runs the other way
The WHO classification does not change at sixty-five, and this page does not pretend it does — no health body has published a replacement band for older adults, and inventing one would be exactly the sort of thing this rebuild exists to stop. What the page adds instead is a reference point, with its evidence attached.
A meta-analysis by Winter and colleagues in the American Journal of Clinical Nutrition in April 2014 pooled 32 studies covering 197,940 community-dwelling adults aged 65 and over, followed for an average of twelve years. Taking a BMI of 23.0 to 23.9 as the reference, mortality was 12 per cent higher at a BMI of 21.0 to 21.9, and 19 per cent higher at 20.0 to 20.9. At the other end, risk did not begin to rise until above a BMI of 33.0. The conclusion was blunt: in older populations being overweight was not associated with increased mortality, while being at the lower end of the recommended adult range was.
The European Society for Clinical Nutrition and Metabolism reaches the practical version of the same conclusion. Its guideline on clinical nutrition and hydration in geriatrics, published by Volkert and colleagues in Clinical Nutrition in February 2019, treats a BMI below 22 in an older person as a marker of malnutrition risk rather than a healthy achievement.
So from sixty-five the tool shows what BMI 23.0 weighs at your height, and an extra column appears in the chart doing the same at every height. Treat it as a floor worth staying above, not a target to aim at. And treat weight loss you did not plan for as something to raise with a doctor rather than something to be pleased about — in later life it is one of the more useful early signals there is.
The weight chart, taken from the authorised standard rather than typed in
The chart on this page lists healthy weight ranges by height. Every row in it is calculated from the classification the tool is already quoting — BMI 18.5 to 25, or 18.5 to 23 if you have selected the WHO 2004 Asian action point — solved for weight at that height. Nothing is transcribed from a printed table, which means the chart cannot drift away from the band shown two sections above it, and it changes correctly the moment you switch between metric and imperial, or between the two scales.
The standard behind it is the World Health Organization adult classification, adopted with matching cut-offs by the NHLBI in its 1998 clinical guidelines and used by the NHS and the CDC. That is the authorised chart. The four formulas are not a standard and never were — they are historical equations from 1964 to 1983, built for drug dosing and diet prescriptions. Where the chart and the formulas disagree, the chart is the answer to use.
Older height-and-weight tables still circulate, most famously the Metropolitan Life Insurance tables of 1959 and 1983, which are organised by height and frame size. They were derived from insured policyholders rather than from health outcomes in the general population, and they were never adopted as a clinical standard. They are historically interesting, and they are not what this chart is based on.
What this page will not do
It will not tell you what you should weigh. It shows you what four historical formulas say, what the current clinical BMI classification implies, and how far apart those answers sit — so you can see that “ideal weight” is a much softer idea than the phrase suggests. Nothing here is medical advice or a treatment plan, and a target weight is a poor primary goal compared with what you eat, how you move and how you sleep.
Nothing is sent anywhere
Every calculation runs in your browser. Height, weight, sex and wrist measurement are never transmitted, stored or logged, and the page keeps working with the connection off once it has loaded.
Questions people ask about ideal body weight
Which of the four formulas is the most accurate?
None of them is accurate, because none was validated against health outcomes. Robinson and Miller are refinements of Devine fitted to newer reference data, so they are arguably better calibrated, but the honest answer is that the healthy BMI range is the better-supported figure and the formulas are historical landmarks.
Why do the formulas give different answers for the same height?
Because they were fitted to different populations at different times for different purposes — drug dosing, diabetes calorie prescriptions, insurance tables. The spread between them at your height is shown for exactly this reason.
What should a 5 ft 4 in woman weigh?
The healthy BMI band runs from roughly 47.8 kg to 64.3 kg (105 to 142 lb). The four formulas cluster in the low-to-mid fifties. The band is the more defensible answer; the formulas sit inside it.
My ideal weight looks very low. Is that normal?
It is common, particularly for men, because Devine and its descendants were estimating lean mass for drug dosing rather than a healthy total weight. If the formula figure sits below the bottom of your healthy BMI range, trust the range.
Does frame size really change the target?
By roughly ten per cent either way, which at average adult height is around six kilograms — not nothing. It is one reason a single number per height was always insufficient.
Should I use the Asian BMI upper bound?
If you are of South, East or South-East Asian descent it is worth looking at, since it reflects that risk rises at a lower BMI in those populations. It is a public-health action point rather than a redefinition of healthy, and it is worth discussing with a doctor rather than adopting silently.
Why does the page show my BMI as well?
Because it had your height and your weight all along and never showed it. Your actual BMI is more informative than a target weight and puts the target in context.
Is there an ideal weight for athletes?
Not from this page. Height-based formulas systematically underestimate appropriate weight for trained people. Body composition is the more useful measure — the body fat calculator answers the question these formulas cannot.
Does age change my ideal weight?
Not in the four formulas — none of them has an age term. It changes which standard applies: below twenty, adult standards do not apply at all; from sixty-five, the evidence says the lower end of the adult band is the risky end rather than the safe one.
Why does the page refuse to answer for a teenager?
Because the adult classification begins at twenty and all four formulas were fitted to adults. Under twenty, weight is assessed on BMI-for-age percentiles against the CDC 2000 growth charts or the WHO 2007 reference for ages 5 to 19, where the healthy band is the 5th to 85th percentile and moves as a young person grows.
What is a healthy weight for a 70-year-old?
The WHO band still applies, but the evidence suggests the bottom of it is not where you want to be. Mortality in adults over 65 rises below a BMI of 23.0 and does not rise at the top end until above 33.0. The tool shows what BMI 23 weighs at your height, as a floor rather than a target.
Is the weight chart on this page an official one?
The ranges are, in the sense that matters: every row is the WHO adult classification solved for weight at that height — the same standard the NHLBI adopted in 1998 and the NHS and CDC use. It is computed rather than copied, so it always matches the band the tool is quoting.
Why not use age-adjusted BMI ranges?
Because they are not current. The 1990 Dietary Guidelines for Americans did publish them, and the 1995 revision withdrew them. Presenting retired guidance as though it were the standard is the opposite of what this page is for.
Do you save what I enter?
No. Nothing is stored or transmitted.
Sources behind these numbers
- Devine BJ, 1974 — the original formula, published for gentamicin dosing, not as a health target.
- Robinson JD et al, 1983, and Miller DR et al, 1983 — two independent revisions of Devine fitted to different reference data.
- Hamwi GJ, 1964 — from diabetes practice, published with the frame-size adjustment this tool restores.
- WHO Expert Consultation, “Appropriate body-mass index for Asian populations”, The Lancet, 2004 (PMID 14726171) — the 23.0 alternative upper bound.
- WHO — obesity, for the classification behind the healthy BMI range.
- NHLBI, National Institutes of Health — Calculate Your BMI.
- CDC — Body Mass Index.
- Winter JE, MacInnis RJ, Wattanapenpaiboon N, Nowson CA, “BMI and all-cause mortality in older adults: a meta-analysis”, American Journal of Clinical Nutrition, April 2014 (PMID 24452240) — 32 studies, 197,940 adults aged 65 and over, mean follow-up 12 years.
- Volkert D et al, “ESPEN guideline on clinical nutrition and hydration in geriatrics”, Clinical Nutrition, February 2019 (PMID 30005900) — the BMI 22 malnutrition risk marker in older people.
- WHO — growth reference data for 5 to 19 years, 2007, for anyone below adult age.
Tools people open next to this one
- Calorie Calculator — the intake side, with safety floors on any deficit.
- Pregnancy Weight Gain Calculator — the right framework if you are pregnant.
- Water Intake Calculator — for the daily fluid figure people usually look up next.