BMR Calculator
Basal metabolic rate is the energy a body spends doing nothing at all — breathing, pumping blood, holding a temperature, replacing cells. Enter your height, weight, age and sex and this runs all three published equations at once, so you can see how far apart they land before you trust any one of them.
Body and equation
This picks the headline figure only. All three are shown below whichever you choose.
Two of the equations were fitted to two groups and carry a different constant for each. Katch-McArdle has no sex term at all.
Whole years. Mifflin-St Jeor subtracts 5 kcal per year, so a decade costs about 50 kcal a day.
Centimetres, without shoes. 1.75 m is 175.
Kilograms. Weigh yourself at the same time of day for a figure you can compare week to week.
Leave this blank unless you have had it measured. A guessed figure makes Katch-McArdle worse than the equations that ignore body composition altogether.
Mifflin-St Jeor predicts 1,669 kilocalories a day at rest. These equations are accurate to roughly ±10% for one person, so read this as 1,502 to 1,836 rather than as a measurement.
- Equation used
- Mifflin-St Jeor
- Likely range for one person
- 1,502 to 1,836 kcal/day
- Spread across the equations
- 53 kcal/day
- Lean body mass
- Add a body fat % to see it
- Sedentary to extra active
- 1,168 kcal/day apart on the table below
| Equation | Estimate | Against the headline |
|---|---|---|
| Mifflin-St Jeor | 1,669 kcal/day | The one shown above |
| Revised Harris-Benedict | 1,722 kcal/day | +53 kcal/day |
| Katch-McArdle | Needs a body fat % | — |
| Activity level | Multiplier | Daily calories |
|---|---|---|
| Sedentary | 1.20 | 2,003 kcal/day |
| Lightly active | 1.375 | 2,295 kcal/day |
| Moderately active | 1.55 | 2,587 kcal/day |
| Very active | 1.725 | 2,879 kcal/day |
| Extra active | 1.90 | 3,171 kcal/day |
This is general information, not medical advice
This page exists to do arithmetic and to explain where the arithmetic came from. It is general educational information about three published prediction equations. It is not medical advice, it is not a diagnosis, and it cannot take account of anything about you that a height, a weight, an age and a sex do not capture — which is nearly everything that moves a metabolic rate.
If you are trying to decide how much to eat, whether your metabolism is unusually slow, or what to do about a weight that is not moving, talk to a doctor or a registered dietitian. They can look at the things this page cannot: your thyroid function, your medications, your history of dieting, your bloods, and whether a prediction equation is the right tool at all. A measured rate, from indirect calorimetry, is available through many clinics and settles the question this page can only estimate.
If you are pregnant or breastfeeding, under 18, an older adult, recovering from illness or surgery, or living with an eating disorder, none of these equations was fitted with you in mind and all of them are especially likely to mislead. Please do not use this page as a reason to eat less.
The three equations, in full
All three take weight in kilograms and height in centimetres. Imperial input is converted once, with the exact definitions — 1 inch = 2.54 cm and 1 pound = 0.45359237 kg — before any coefficient is applied.
Mifflin-St Jeor: 10W + 6.25H − 5A + 5 (men) or − 161 (women)- Revised Harris-Benedict, men
- 88.362 + 13.397W + 4.799H − 5.677A
- Revised Harris-Benedict, women
- 447.593 + 9.247W + 3.098H − 4.330A
- Katch-McArdle
- 370 + 21.6 × lean body mass in kg
- W, H, A
- Weight in kg, height in cm, age in years
Katch-McArdle is the odd one out in three ways: it has no sex term, no age term, and it cannot run at all without a body fat percentage. That is the design, not an omission. Its claim is that the differences the other two equations use sex and age to approximate are really differences in lean tissue, so a direct measurement of lean tissue makes both terms redundant. When no body fat figure is supplied, this calculator leaves the row empty rather than substituting an assumption.
The 370 and 21.6 in that row are not a measurement Katch and McArdle made themselves. They match, coefficient for coefficient, the general prediction equation Cunningham published in 1991 — a pooled re-analysis of previously reported resting-expenditure studies in adult men and women, not a new sample. Katch and McArdle’s own textbooks popularized the equation under their names, which is why it carries them here too; the sources below cite Cunningham’s paper instead, because it is the one with a method section, a population, and a DOI.
Every figure is rounded to whole kilocalories. A tenth of a kilocalorie on a number carrying a ±10% individual error would be four digits of precision on one digit of accuracy.
The result panel also runs the headline figure through five standard activity multipliers — sedentary at 1.2 up to extra active at 1.9 — as a reference table. These are not a fourth equation: they are a naming convention repeated in fitness material since the 1990s, the same five figures the TDEE calculator applies, and no study fitted them the way Mifflin-St Jeor or Cunningham’s equation were fitted. They are shown unselected, for the same reason the three equations are shown together — so the spread is visible before any one number is trusted.
A 30-year-old man, 175 cm and 72 kg, three ways
Mifflin-St Jeor. 10 × 72 = 720. 6.25 × 175 = 1093.75. 5 × 30 = 150. So 720 + 1093.75 − 150 + 5 = 1668.75, which prints as 1,669 kcal a day.
Revised Harris-Benedict. 13.397 × 72 = 964.58. 4.799 × 175 = 839.83. 5.677 × 30 = 170.31. So 88.362 + 964.58 + 839.83 − 170.31 = 1722.46, or 1,722 kcal — 53 more than Mifflin-St Jeor for the identical body.
Katch-McArdle. This one needs a fourth number. At a measured 20% body fat, 72 kg is 57.6 kg of lean mass, so 370 + 21.6 × 57.6 = 1614.16, or 1,614 kcal — 108 below the Harris-Benedict figure and 55 below Mifflin.
Three equations, one body, a 108 kcal spread. Now apply the ±10% individual error that any of them carries: the Mifflin figure of 1,669 is really 1,502 to 1,836. The spread between the equations is smaller than the uncertainty inside each of them, which is the most useful thing on this page.
The activity table, one row of it. Take the Mifflin-St Jeor figure into the reference table below. At sedentary, 1,669 × 1.2 = 2,003 kcal a day. At extra active, 1,669 × 1.9 = 3,171. That 1,168 kcal gap is ten times the 108 kcal spread between all three equations above — the reason this page shows every level rather than asking which word describes your week.
These are regression lines, not measurements
A real basal metabolic rate is measured, not calculated. The method is indirect calorimetry: a fasted person lies still under a hood or a mask while the oxygen they consume and the carbon dioxide they produce are analysed, and their energy expenditure is worked out from the gas exchange. What these equations do is take a few hundred such measurements, fit a line through them against height, weight, age and sex, and then hand that line to somebody who was not in the sample.
That method has three consequences worth stating plainly.
- The individual error is about ±10%.That is not a caveat about the arithmetic; it is the width of the scatter around the fitted line. Roza and Shizgal put Harris-Benedict’s precision at 14%. Mifflin-St Jeor is better, but not by an order of magnitude. Two people with identical height, weight, age and sex can genuinely differ by 300 kcal a day, mostly in organ mass and lean tissue, and no equation here has a term for it.
- The samples were small and local.Mifflin-St Jeor was fitted to 498 healthy adults — 247 women and 251 men, aged 19 to 78, including both normal-weight and obese participants — recruited in the United States and measured in one laboratory. Harris and Benedict measured 239 people in Boston, published in 1919; the 1984 Roza-Shizgal revision refit those data together with 98 further subjects. Katch-McArdle is different again: its real source, Cunningham’s 1991 paper, is not one cohort but a synthesis of other researchers’ studies of resting expenditure in adult men and women — a different kind of uncertainty than a single sample carries, not a smaller one. None of the three equations has a term for ancestry, and none of the samples was globally representative.
- They cannot see anything they were not given. Thyroid function, medication, pregnancy, illness, fever, recent severe restriction, sleep, and the sheer variation in organ mass between two similar-looking people all move a resting rate, and none of them appears in a formula with four inputs.
That does not make the number useless. It makes it a starting estimate with a stated error, which is a genuinely useful thing to have — provided you carry the error along with it rather than dropping it the moment the four digits appear.
BMR is not the number most people want
Almost nobody spends the day lying still. The figure people usually mean by “maintenance calories” is total daily energy expenditure — this resting rate multiplied by an activity factor between 1.2 and 1.9. The reference table above turns that into five numbers for this body; picking one, comparing it against your own guess, and reading the multiplier’s own honest limits is what the TDEE calculator does.
If you want the Katch-McArdle column filled in and have never had your body composition measured, the body fat calculator runs the US Navy circumference method, which needs a tape measure rather than a laboratory. It is an estimate too — but a measured circumference is a much better input than a guess.
What this calculator assumes
- You are a healthy adult. The Mifflin-St Jeor sample ran from 19 to 78 years old, and none of these equations was fitted to children, adolescents, or people who were acutely ill.
- You are not pregnant or breastfeeding. Both raise energy requirements by amounts these equations have no term for.
- Nothing is altering your metabolic rate pharmacologically or clinically — no untreated thyroid disorder, no medication that shifts resting expenditure.
- Any body fat percentage you enter was measured rather than estimated. Katch-McArdle multiplies your lean mass by 21.6, so it multiplies the error in that figure too.
- “BMR” and “RMR” are being used interchangeably, as they are almost everywhere. Measured strictly, resting rate runs about 10% above basal rate, and the three equations here were not all fitted to the same one.
- No activity is included. This is the rest-only figure; walking, working and training are added by an activity multiplier, not by these equations.
- Imperial figures are converted with the exact definitions — 1 inch = 2.54 cm, 1 pound = 0.45359237 kg — so the two unit systems agree to within a rounding step.
- The activity-multiplier reference table repeats the same five figures the TDEE calculator applies to this exact bmr. It adds no term for the thermic effect of food or non-exercise activity, and turning any of its five rows into an eating plan is a separate decision with its own honest limits.
- The Katch-McArdle coefficients — 370 and 21.6 per kilogram of lean mass — come from Cunningham’s 1991 pooled re-analysis of earlier studies, not from a new sample Katch and McArdle measured themselves.
BMR FAQ
Which BMR formula is the most accurate?
Mifflin-St Jeor, for the general population. The Academy of Nutrition and Dietetics commissioned a systematic review of resting metabolic rate prediction — Frankenfield, Roth-Yousey and Compher, 2005 — which compared the published equations against measured values and found Mifflin-St Jeor the most reliable in both non-obese and obese adults. That is why it is the default here. Katch-McArdle can beat it, but only for someone who has had their body fat properly measured; fed a guess, it is the worst of the three, because a wrong lean-mass figure is multiplied by 21.6.
Is BMR the same as RMR?
Not strictly, though almost everyone uses the terms interchangeably and this page does too. Basal metabolic rate is measured under laboratory conditions: an overnight fast, complete rest, a thermally neutral room, and often on waking. Resting metabolic rate is measured under looser conditions and comes out roughly 10% higher. The awkward part is that the equations mix the two — Harris and Benedict measured basal rate, Mifflin and colleagues measured resting energy expenditure — so the label on the output is less precise than the four-digit number suggests.
Should I eat my BMR?
That is a question for a clinician, not a calculator, and this page is not the place to get an answer. What can be said factually is that BMR is what a body spends lying still and doing nothing — it excludes walking, working, cooking, fidgeting and training, all of which are real and add up. Eating at BMR is therefore a substantial deficit for almost anyone who gets out of bed. The maintenance figure people usually mean is total daily energy expenditure, which is BMR multiplied by an activity factor.
Why does another calculator give me a different BMR?
Almost always because it uses a different equation, and sometimes because it uses the original 1919 Harris-Benedict coefficients rather than the 1984 Roza-Shizgal revision. On the worked example below, Mifflin-St Jeor and revised Harris-Benedict differ by 53 kcal a day on identical inputs, and adding Katch-McArdle widens the spread to 108. None of them is wrong; they are three regression lines through three different samples, and they disagree because people do.
Do I need my body fat percentage?
Only if you want the Katch-McArdle figure, and only if the percentage is measured rather than estimated. DXA, hydrostatic weighing, air displacement and a careful skinfold or circumference protocol all produce something usable. A number from a bathroom scale's bioimpedance reading, or an eyeballed guess from a photo comparison, does not — the error in the body fat figure is amplified by the equation, so a bad input makes the result worse than the equations that ignore body composition entirely.
Does BMR fall as you lose weight?
Yes, in two ways, and the equations only capture one of them. The obvious one is arithmetic: every equation here has a weight term, so a lighter body predicts a lower rate. The less obvious one is adaptive — during and after sustained energy restriction, measured resting expenditure tends to fall by more than the change in body mass alone would predict. No prediction equation has a term for how you arrived at today's weight, so a figure calculated after a long diet is likely to read high.
Why doesn't the BMR calculator let me pick an activity level?
Because BMR and an activity guess carry very different amounts of error, and folding them into one number would hide that difference. The equations above are accurate to roughly ±10% for one person; the activity multiplier is a self-assessment chosen from five buckets fifteen percentage points apart, which is a far bigger and far less disciplined source of error. This page shows the reference table — what the headline figure becomes at each of the five levels — without asking you to commit to one. The TDEE calculator is where you actually pick a level, compare it against the others, and read about why the multiplier is the weakest link in the whole calculation.
Where do the activity multipliers 1.2 through 1.9 come from?
Nowhere as rigorous as the three equations above, and that is worth saying plainly. They are a naming convention that has circulated in fitness material since the 1990s — sedentary at 1.2, extra active at 1.9 — not coefficients fitted to a measured sample the way Mifflin-St Jeor or Cunningham's equation were. They are also not the same scale as the physical activity level (PAL) values in the FAO/WHO/UNU human energy requirements report, which run from 1.40 to 2.40 for the same idea. Every level is shown here, rather than one being recommended, because there is no published evidence to prefer one over another.
Is the Katch-McArdle formula really Katch and McArdle's own research?
The name is theirs — it comes from their exercise physiology textbooks, which is where most readers encounter it. The coefficients are not: 370 + 21.6 × lean body mass is, term for term, the general prediction equation Cunningham published in 1991 after pooling previously reported studies of resting expenditure in adult men and women. This page keeps the popular name, because that is what people search for, and cites Cunningham's paper in the sources below, because it is the one with a method section, a population, and a DOI.
Sources and review notes
- Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. “A new predictive equation for resting energy expenditure in healthy individuals.” American Journal of Clinical Nutrition, 1990;51(2):241–247 — the source of the default equation and of the 498-person sample described above
- Frankenfield D, Roth-Yousey L, Compher C. “Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review.” Journal of the American Dietetic Association, 2005;105(5):775–789 — the Academy of Nutrition and Dietetics review that put Mifflin-St Jeor ahead of the alternatives
- Roza AM, Shizgal HM. “The Harris Benedict equation reevaluated: resting energy requirements and the body cell mass.” American Journal of Clinical Nutrition, 1984;40(1):168–182 — the 1984 refit used here, and the source of the 14% precision figure
- Academy of Nutrition and Dietetics — the professional body behind the 2005 review, and the place to find a registered dietitian if you want the question answered properly
- Cunningham JJ. “Body composition as a determinant of energy expenditure: a synthetic review and a proposed general prediction equation.” American Journal of Clinical Nutrition, 1991;54(6):963–969 — the 370 + 21.6 × lean mass equation this page labels Katch-McArdle, and the pooled studies of adult men and women it was fitted to
- Human energy requirements: report of a Joint FAO/WHO/UNU Expert Consultation, Rome 2001 — the factorial method behind the activity-multiplier reference table, and the 1.40-to-2.40 physical activity level scale this page’s 1.2-to-1.9 ladder is not the same as
The coefficients on this page are fixed constants from published papers and do not go stale. What does change is the evidence about which equation to prefer, and for which people; where a newer review or your own clinician disagrees with the ordering here, take theirs. The five activity multipliers in the reference table are the exception — a convention rather than a finding, which is why every level is shown rather than one being recommended.