BMR Formula Comparison Calculator

Why so many BMR equations?

Basal metabolic rate (BMR), the energy your body burns at complete rest just to stay alive, accounts for 60-75% of total daily energy expenditure in most adults. Because BMR is so central to nutrition planning, researchers have spent more than a century trying to predict it from easily measured variables. The result is a family of equations, each developed on a different population using a different statistical method, each with its own strengths and weaknesses. This calculator lets you see all six major equations side-by-side so you can choose the one that fits your situation best.

The six major BMR equations

Mifflin-St Jeor (1990). The current recommendation of the Academy of Nutrition and Dietetics for healthy adults. Men: BMR = 10×kg + 6.25×cm − 5×age + 5; Women: BMR = 10×kg + 6.25×cm − 5×age − 161. Developed on 498 healthy adults including obese individuals, it tends to be the most accurate for the general modern population and typically predicts lower than the older Harris-Benedict equations.

Harris-Benedict Original (1919). The first widely adopted BMR equation, developed on 239 subjects. Men: 66.5 + 13.75×kg + 5.003×cm − 6.75×age; Women: 655.1 + 9.563×kg + 1.850×cm − 4.676×age. Tends to over-predict modern BMR by 5-15% due to differences in the 1919 cohort.

Harris-Benedict Revised (1984). Roza and Shizgal's update using a larger and more modern database. Men: 88.362 + 13.397×kg + 4.799×cm − 5.677×age; Women: 447.593 + 9.247×kg + 3.098×cm − 4.330×age. Closer to modern BMR than the 1919 version but still slightly higher than Mifflin.

Katch-McArdle (1977). The first widely used LBM-based equation. BMR = 370 + 21.6×LBM(kg) for both sexes. Best for athletes, bodybuilders, and lean individuals because it directly accounts for body composition. Requires body fat percentage.

Cunningham (1991). Derived from 226 trained athletes. BMR = 500 + 22×LBM(kg). Predicts 3-8% higher than Katch-McArdle, reflecting the elevated muscle metabolic rate that comes from chronic training. Use for serious athletes (5+ hours/week for 6+ months).

Owen (1986). Developed on 44 lean and 67 obese women, then extended to men. Men: 879 + 10.2×kg; Women: 795 + 7.18×kg. Uses only weight, no height or age. Particularly well-validated in lean and obese women, where other equations may mis-estimate.

How the equations differ in practice

For a typical 75 kg, 178 cm, 30-year-old man with 15% body fat (LBM ≈ 63.75 kg), the six equations give: Mifflin ≈ 1720, HB Original ≈ 1819, HB Revised ≈ 1775, Katch-McArdle ≈ 1747, Cunningham ≈ 1903, Owen ≈ 1644 kcal/day. The spread is 259 kcal, a 14% range that, if applied naively to a 12-week weight loss plan, would compound into a 6-7 pound difference in expected weight loss. This is why choosing the right equation matters.

Which equation should you use?

For general adults without body composition data, use Mifflin-St Jeor. It's the current evidence-based recommendation and performs best across diverse populations. If you know your body fat percentage and you're an athlete or very lean, use Katch-McArdle (or Cunningham if you're a trained athlete). If you're a woman with very low or very high body weight, Owen may give a more accurate estimate. The Harris-Benedict equations are useful primarily for historical continuity with older research protocols.

Why the equations disagree

The disagreement comes from three sources. First, the underlying populations differ. Harris-Benedict's 1919 cohort was small and predominantly white; Mifflin-St Jeor's 1990 cohort was larger and more diverse; Cunningham's 1991 sample was specifically trained athletes. Second, the statistical methods differ. Harris-Benedict used simple linear regression on log-transformed data; Mifflin used multiple regression on raw values. Third, the predictor variables differ, weight-based equations (Harris-Benedict, Mifflin, Owen) cannot distinguish a muscular 80 kg person from a sedentary 80 kg person, while LBM-based equations (Katch-McArdle, Cunningham) can.

The body composition problem

The biggest source of error in BMR prediction is body composition. A 100 kg sedentary individual and a 100 kg bodybuilder have very different resting energy expenditures, but weight-based equations give them identical estimates. The Katch-McArdle and Cunningham equations solve this by using LBM, but they require an accurate body fat measurement, which is itself difficult. The most accurate body fat methods (DEXA, hydrostatic weighing) are expensive and require clinic visits; the most accessible methods (BIA scales, skinfold calipers) can be off by 4-8 percentage points. If your body fat measurement is unreliable, you may be better off with Mifflin-St Jeor than with a poorly-measured LBM.

From BMR to TDEE

Once you've chosen your BMR estimate, multiply by an activity factor to get total daily energy expenditure (TDEE). Standard factors are: 1.2 (sedentary), 1.375 (light activity), 1.55 (moderate activity), 1.725 (very active), and 1.9 (extra active). These factors come from the original Harris-Benedict paper and remain widely used. For athletes in heavy training, factors of 1.9-2.4 may be more appropriate. The factor you choose has a larger impact on TDEE than the BMR equation you choose, a difference of 0.2 in the activity factor (e.g. 1.55 vs 1.75) shifts TDEE by roughly 300-400 kcal for most adults.

How to validate your chosen equation

The ultimate test is real-world outcomes. Track your weight daily for 2-4 weeks while logging your calorie intake. If your weight stays stable, your actual TDEE equals your intake. If you gain or lose, your actual TDEE is lower or higher than your intake by approximately 3500 kcal per pound of weight change per week. Adjust your BMR-based estimate accordingly. This calibration process is more valuable than choosing the "right" equation, because individual variation in metabolic rate can be ±10-15% even with the best equation.

The mathematical structure of BMR equations

All BMR equations share a common mathematical structure: a linear (or polynomial) combination of anthropometric variables, with coefficients fit by regression. The Mifflin-St Jeor equation uses the form BMR = a×weight + b×height + c×age + d×sex, where sex contributes a constant (+5 for men, −161 for women). The Harris-Benedict equations use the same form but with different coefficients derived from earlier data. The Katch-McArdle and Cunningham equations use a different structure: BMR = constant + coefficient × LBM, where LBM is itself derived from weight and body fat percentage. The Owen equations use the simplest form: BMR = constant + coefficient × weight, with no height or age terms.

The choice of predictor variables reflects different assumptions about what drives metabolic rate. Weight-only equations (Owen) assume that total weight captures most of the variance in BMR, true on average but inaccurate for individuals with non-average body composition. Weight + height + age equations (Mifflin, Harris-Benedict) capture more variance because height correlates with frame size and age captures the metabolic decline with aging. LBM-based equations (Katch-McArdle, Cunningham) capture the most variance because lean mass is the primary driver of BMR, but they require body fat measurement, which adds cost and potential error. No equation captures all variance; the residual 10-20% is due to genetic factors, thyroid function, brown fat activity, and other individual characteristics that no anthropometric equation can predict.

A practical decision tree for choosing an equation

Use this decision tree to choose the right equation for your situation. Step 1: Do you have a reliable body fat measurement (within 3-4 percentage points)? If no, use Mifflin-St Jeor, it's the most accurate equation that doesn't require body composition. Step 2: If yes, are you a trained athlete (5+ hours/week for 6+ months)? If yes, use Cunningham, it captures the elevated muscle metabolic rate from chronic training. Step 3: If not an athlete but you have body composition data, use Katch-McArdle, it's more accurate than weight-based equations for any individual with known body composition. Step 4: If you're a woman with very low or very high body weight, consider Owen, it was specifically validated in lean and obese women. Step 5: If you're working with older research data or military nutrition planning, use Harris-Benedict (1919 or 1984) for continuity.

This decision tree gives you the best equation for your situation, but remember that the difference between equations is typically 100-300 kcal/day, much smaller than the variation introduced by activity factor choice (which can shift TDEE by 400-800 kcal/day). Spend more time accurately estimating your activity level than agonizing over which BMR equation to use. And whatever equation you choose, validate it against real-world weight tracking over 2-4 weeks of consistent intake, then adjust as needed.

The role of the thermic effect of food

Beyond BMR and activity, the thermic effect of food (TEF) contributes 5-15% of total daily energy expenditure. TEF is the energy cost of digesting, absorbing, and storing the nutrients you eat. Protein has the highest TEF (20-30% of calories consumed), followed by carbohydrates (5-10%), fat (0-3%), and alcohol (10-30%). A mixed diet typically has a TEF of about 10% of total calorie intake, so if you eat 2,500 kcal/day, TEF contributes about 250 kcal to daily expenditure. TEF is not captured by any BMR equation, but it's a meaningful contributor to total energy balance. High-protein diets slightly raise TDEE through TEF, about 50-100 kcal/day for a 1,500 kcal high-protein diet vs a high-carb diet of the same caloric value. This is one reason high-protein diets are slightly more effective for weight loss than calorie-matched high-carb diets.