Which BMR formula is most accurate?

Mifflin-St Jeor (1990) is the default here because a 2005 systematic review of the four most commonly used equations found it predicted resting metabolic rate within 10% of measured values in more people than any other equation compared, with the narrowest error range. That is a statement about groups of people, not a promise for any one person, and the review noted that older adults and US ethnic minorities were underrepresented in the research. If you know your body fat percentage, Katch-McArdle works from lean body mass instead of assuming a typical body composition for your height and weight, which can suit people who are leaner or more muscular than average.

Why do the three formulas give different numbers?

They were built from different research cohorts at different times, using different inputs. Harris-Benedict dates to 1919 (revised in 1984) and doesn't use body fat at all. Mifflin-St Jeor (1990) uses the same basic inputs but was fitted to a more contemporary population. Katch-McArdle (1996) drops total weight in favour of lean mass. None is 'wrong' — they're different models of the same thing, built on different assumptions.

Can BMR be measured directly instead of estimated?

Yes, via indirect calorimetry — measuring the oxygen you consume and carbon dioxide you produce at rest, typically in a clinical or research setting. It is far more accurate than any formula but not something most people have access to, which is why equation-based estimates like these remain the practical default.

Does BMR change with age or muscle mass?

Yes to both. BMR tends to decline gradually with age, partly because lean body mass tends to decrease over time. More muscle mass raises BMR, because muscle tissue burns more energy at rest than fat tissue does — part of why Katch-McArdle, which uses lean mass directly, tends to suit muscular individuals better than formulas based on total weight alone.

Is BMR the same as the calories I should eat?

No — BMR is only the resting component. Almost everyone should eat more than their BMR, since it excludes the energy used by movement, exercise, and digestion. For a full daily target, multiply BMR by an activity factor to get TDEE — see the TDEE calculator.

BMR Calculator — Mifflin-St Jeor vs. Harris-Benedict Formula

Formula	BMR
Mifflin-St Jeor	1370 kcal
Harris-Benedict (revised)	1430 kcal
Katch-McArdle	Add body fat % to compare

What BMR measures

Basal metabolic rate is the energy your body burns at complete rest — enough to keep your heart beating, lungs breathing, and organs functioning, with no digestion, movement, or exercise included. For most people it accounts for the majority of total daily energy expenditure, which is exactly why it's the starting point every calculator on this site builds from.

Three formulas, three eras

Harris-Benedict is the oldest of the three, originally published in 1919 and revised by Roza and Shizgal in 1984. It estimates BMR from sex, age, height, and weight, and was the standard for decades. Mifflin-St Jeor, published in 1990, uses the same four inputs but was fitted to a more contemporary population. A 2005 systematic review found it predicted measured resting metabolism within 10% for more people than the other equations it compared, which is why it's the default formula across this site. Katch-McArdle, from 1996, takes a different approach entirely: instead of total weight, it uses lean body mass, calculated from your weight and body fat percentage. That can make it a better fit for people who are leaner or more muscular than average for their height, since it isn't assuming a typical ratio of fat to muscle.

The formulas as used on this page

Here are the equations behind the numbers above, with weight in kilograms, height in centimetres, and age in years. The Mifflin-St Jeor equations are as printed in the 1990 paper's abstract; the other two are as implemented in the calculator on this page.

Mifflin-St Jeor (1990): men, 10 × weight + 6.25 × height − 5 × age + 5; women, 10 × weight + 6.25 × height − 5 × age − 161.
Harris-Benedict, revised (1984): men, 88.362 + 13.397 × weight + 4.799 × height − 5.677 × age; women, 447.593 + 9.247 × weight + 3.098 × height − 4.33 × age.
Katch-McArdle (1996): 370 + 21.6 × lean body mass, where lean body mass is your weight × (1 − body fat percentage).

"Harris-Benedict" on this site always means the 1984 revision by Roza and Shizgal. The original 1919 coefficients are not used anywhere here.

How to read the comparison

The three numbers won't match exactly, and that's expected — they're different models, not different calculations of the same fixed truth. If you don't know your body fat percentage, compare Mifflin-St Jeor and Harris-Benedict and lean toward Mifflin-St Jeor as the generally better-validated of the two. If you do know your body fat percentage and it's notably above or below what's typical for your build, Katch-McArdle is worth weighting more heavily.

A worked example

Take a 35-year-old woman, 165cm tall, weighing 70kg. Mifflin-St Jeor puts her BMR at about 1,395 kcal. The revised Harris-Benedict equation, using the same four inputs, comes out a little higher at about 1,455 kcal — the two formulas were fitted to different research cohorts, so a gap of this size is normal rather than a sign one of them is wrong. Now suppose she knows her body fat is 32%: Katch-McArdle works from her lean mass alone (about 47.6kg) and produces a third figure, about 1,398 kcal, independent of the population assumptions baked into the other two. None of the three is "the" right answer — they're three reasonable estimates, and the spread between them is a useful signal of how much uncertainty sits underneath any single-number BMR claim.

Who this suits

Anyone who wants to see the resting number on its own — before an activity multiplier, a goal, or a macro split gets applied — or who wants to sanity-check one formula against another rather than trust a single output. It's also useful if you're comparing this calculator's numbers against another app or a clinician's estimate and want to understand why they might differ.

Limitations

All three formulas are equations fitted to research populations, not a measurement of your actual metabolism. The only way to measure BMR directly is indirect calorimetry — tracking oxygen consumption and carbon dioxide output at rest, usually in a lab. Formula-based estimates are a practical substitute, but individual errors can be real. The 2005 review concluded that Mifflin-St Jeor is more likely than the other equations tested to land within 10% of measured resting metabolism, but that noteworthy errors and limitations exist when it is applied to individuals. Roza and Shizgal reported a precision of ±14% for the Harris-Benedict equations in normally nourished people, and found them unreliable in malnourished patients. Individual factors like thyroid function, medication, or genetics may also shift real resting energy expenditure up or down in ways an equation can't see. Anyone with a medical question about their metabolism should ask a doctor or dietitian.

Sources

Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr. 1990;51(2):241-247. PubMed
Roza AM, Shizgal HM. The Harris Benedict equation reevaluated: resting energy requirements and the body cell mass. Am J Clin Nutr. 1984;40(1):168-182. PubMed
Frankenfield D, Roth-Yousey L, Compher C. Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review. J Am Diet Assoc. 2005;105(5):775-789. PubMed

What Is TDEE? Total Daily Energy Expenditure — how BMR fits into the total you burn in a day.
How Accurate Are Calorie Counts, Really? — the other half of the uncertainty: how much error sits in the food numbers too.
How Much Protein Per Meal Do Older Adults Need — why BMR trending down with age changes the protein math too.

BMR Calculator