Four Ideal Weight Formulas, Four Different Answers — and All of Them Were Built for Something Else

A 5'10" man looks up his ideal weight. Devine says 160.9 lb. Robinson says 161. Miller says 156.4. Hamwi says 166. The healthy BMI range for that height spans roughly 129 to 174 pounds — a forty-five pound window.

None of these is wrong. They disagree because they were built to answer different questions, mostly not the question a person standing on a scale is actually asking. Understanding where each came from tells you which one, if any, deserves your attention.

Where the Formulas Actually Come From

Hamwi (1964) was a diabetes dosing shortcut. Developed to estimate insulin requirements and nutritional needs in clinical practice. For men: 106 lb for the first 5 feet, plus 6 lb per additional inch. For women: 100 lb plus 5 lb per inch. Deliberately simple, because it was meant to be done mentally at a bedside.

Devine (1974) was a drug dosing formula. This one is the most widely used and the most widely misunderstood. Devine created it to calculate loading doses of medications that distribute in lean tissue rather than fat — where dosing on total body weight would overdose a heavier patient. It was never validated as a health target. For men: 50 kg + 2.3 kg per inch over 5 feet. For women: 45.5 kg + 2.3 kg per inch.

Robinson (1983) and Miller (1983) were corrections to Devine. Both were attempts to fit the equations more closely to observed population data. Miller produces notably lower values at greater heights, which is why it's usually the outlier in a side-by-side comparison.

The healthy BMI range is a population statistic. BMI of 18.5–24.9 came from epidemiological associations between weight-for-height and mortality across large groups. It describes where risk is lowest on average in a population. It says nothing about any individual, and it doesn't distinguish muscle from fat.

So of the four "ideal weight" formulas in common use, one was for insulin, one was for drug dosing, and two were adjustments to a drug dosing formula. The ideal weight calculator will run all of them side by side, and the spread you see is the honest answer.

Why They Disagree So Much

All of them use height alone. Frame size, bone density, limb proportions and muscle mass are absent from every one of these equations. A 5'10" person with broad shoulders and heavy bones and a 5'10" person with a slight frame get identical targets, which is obviously incorrect and unavoidable given the inputs.

They're linear; bodies aren't. Each formula adds a fixed number of pounds per inch of height, which fits reasonably near the middle of the population distribution and progressively worse at the extremes. This is why the formulas diverge most for very tall and very short people — exactly the individuals for whom a target would be most useful.

They were derived from specific, narrow populations. Mid-twentieth-century clinical samples, largely of one ethnic background. Body composition norms differ meaningfully across ancestries; several health bodies use lower BMI thresholds for populations of South Asian descent for precisely this reason.

They don't know what year it is. Fifty-year-old equations encode fifty-year-old population data and fifty-year-old clinical purposes.

What to Use Instead

The genuinely useful question isn't "what should I weigh" but "is my body composition and fat distribution in a healthy range." Three measurements answer that far better than any height formula.

Waist-to-height ratio. Keep your waist circumference below half your height. It's a single division, requires only a tape measure, and performs at least as well as BMI for predicting cardiometabolic risk in most studies — better in some, because it captures visceral fat, which is the fat that actually matters. A 5'10" (70 inch) person wants a waist under 35 inches.

Body fat percentage. The number the ideal-weight formulas were reaching for indirectly. General health ranges run roughly 10–20% for men and 18–28% for women, with athletic ranges lower and the useful floor higher than most people assume. Estimate it with the body fat calculator and understand the error bars of whichever method you use.

Lean body mass, tracked over time. Two people at the same weight with a ten-pound difference in lean mass are in quite different health positions. The lean body mass calculator gives an estimate; its trend matters more than its absolute value.

And the healthy weight range, as a range. The healthy weight calculator and the BMI calculator will give you a span rather than a point. A span is the honest form of this answer — for 5'10", that's roughly 129–174 lb, and where you sit within it depends on the composition questions above.

Where a Single Number Is Genuinely Appropriate

There are contexts where a point estimate is required and these formulas remain the right tool.

Clinical drug dosing. Devine and its descendants are still standard for calculating doses of medications that distribute in lean tissue. This is the job they were built for and they do it well.

Ventilator tidal volume settings. Predicted body weight, derived from these formulas, is standard practice in critical care.

Nutritional requirement estimates in clinical settings. Where a rapid, standardised figure is needed for a patient whose actual weight may be distorted by fluid retention.

Notice that all three are situations where a clinician needs a defensible standard number quickly for a specific technical calculation — not situations where someone is deciding whether their body is acceptable.

What to Actually Do

Stop looking for the number and establish a range. Take the healthy BMI span for your height, then narrow it using waist-to-height ratio and body fat estimate. If your waist is under half your height and your body fat sits in a normal range, your weight is fine wherever it lands in the span.

Measure your waist and divide by your height. Right now, in thirty seconds. If the result is under 0.5, that single fact is more informative than every formula in this article.

Track direction over months, not position on a chart. Waist circumference falling while strength holds or improves is unambiguous progress, regardless of what any formula says you should weigh.

Treat the formulas as historical artefacts, which they are. They're in every calculator because they've always been in every calculator, not because anyone validated them against the question you're asking.

The reason four formulas give four answers is that none of them was ever designed to tell you what to weigh. Take the tape measure instead.