Ideal Weight Formulas Explained: Devine, Robinson, and Hamwi

If you have ever typed your height into a health app and gotten back one clean "ideal weight" number, you have probably wondered where that number came from. The honest answer is that it came from a formula written down decades ago by a clinician who needed a quick estimate, not a personal verdict about your body. There are several of these formulas, they disagree with each other, and none of them can see anything about you except your height and your sex.

That sounds like a reason to ignore them, but it is actually a reason to understand them. Once you know what each formula assumes, the spread between them becomes useful information instead of noise. This post walks through the three most quoted equations, shows the math on a real height, and explains the two caveats that matter most: frame size and muscle.

Where these formulas come from

The four equations you will see most often are named after the people who published them. Hamwi (1964) was a fast bedside rule for estimating insulin doses. Devine (1974) was built for calculating drug doses by body weight and is still the version pharmacy and renal-dosing software uses today. Robinson (1983) and Miller (1983) reworked Devine's coefficients against their own patient data.

What they share is the same skeleton: a base weight at five feet of height, plus a fixed increment for every inch above that. They differ only in the size of the base and the increment, which is why they spread apart as you get taller. They were all fitted on adults five feet and up, in the 1960s through 1980s, which is worth keeping in mind before treating any of them as a modern standard.

The Devine formula, step by step

Devine is the cleanest one to read, and it is the formula most clinical forms actually ask for. Here it is:

Devine, men = 50 kg + 2.3 kg for every inch over 5 ft
Devine, women = 45.5 kg + 2.3 kg for every inch over 5 ft

So the only two things the equation needs are your sex (which picks the base) and your height in inches above five feet (which scales the rest). Everything else about you is invisible to it. Robinson and Miller use the same shape with slightly lower numbers, which is why they usually land a kilogram or two under Devine. Hamwi uses pounds in its original form and tends to sit a touch higher for tall men.

A worked example: a 178 cm man

Let me run my own height through it, because seeing the arithmetic makes the limitations obvious. I am 178 cm, which is about 70 inches, or 10 inches over five feet.

Devine for men gives 50 + (2.3 × 10) = 73 kg. Robinson comes out lower, near 70 kg. Miller is lower still, around 68 kg. Hamwi lands close to Devine. So the four formulas hand me a range of roughly 68 to 73 kg, with a consensus average near 71 kg.

Now compare that to the healthy BMI band for 178 cm, which runs from about 58.6 kg to 78.8 kg (BMI 18.5 to 24.9). The formula range sits comfortably inside that band, slightly above the middle. That tells me the formulas and the BMI screen agree, and that anywhere in the upper 60s to low 70s is a defensible place for a frame like mine. It does not tell me that 71 kg is a target I should be chasing to the decimal. You can run this same comparison on the ideal weight calculator, which shows all four formulas side by side with the BMI band drawn around them.

Why frame size changes the answer

Two people who are both 178 cm can have genuinely different healthy weights because of bone structure. A person with wide wrists, broad shoulders, and a deep ribcage carries more weight at the same body fat than someone with a narrow build, and that difference is real, not vanity.

Of the four formulas, only Hamwi's original 1964 version wrote down a correction for this: subtract ten percent for a small frame, add ten percent for a large frame. Devine, Robinson, and Miller published no frame term at all, so a tool that respects the source math leaves them untouched rather than inventing an adjustment. On a 73 kg Hamwi result, that ten percent swing is roughly seven kilograms in either direction, which is large enough to move you across the whole spread the four formulas produced. If you have ever felt that a calculator's number was off for your build, frame size is usually the reason.

Muscle is the blind spot the formulas cannot fix

The bigger limitation is that height-only formulas have no idea what your weight is made of. Muscle is denser than fat, so a trained body weighs more at the same size. A 178 cm rugby player at twelve percent body fat and a 178 cm office worker at twenty-eight percent body fat get the exact same Devine number, even though one of them is lean and the other is not.

This is why the formulas systematically mislabel athletes as "overweight." The math sees a heavy person and has no way to know the weight is lean mass. If you lift or train seriously, a body-composition estimate describes you far better. Run the body fat calculator to see your lean mass, and pair it with the BMR calculator to understand the energy that muscle actually costs to maintain. Those two numbers say more about a trained body than any height-and-sex formula can.

How to actually use the number

So what is the point of an ideal weight figure if it cannot account for your frame or your muscle? Use it as a sanity check, not a destination. The range across the four formulas, sitting inside the wider BMI band, gives you a window. If a goal weight you had in mind falls below the lowest formula result, that is a flag worth discussing with a coach before you chase it. If your current weight is inside the BMI band but above the formula average, that usually just means the formulas are conservative for your build, and nothing is wrong.

Treat the consensus average as the center of a target window, never as a hard number with a decimal point. The formulas are forty to sixty years old, they only know two facts about you, and they disagree by several kilograms among themselves. Read them as a rough guide, layer in your frame and your training, and they become a quick reality check instead of a rule you are failing to meet.

Made by Toolora · Updated 2026-06-13