How the TDEE Calculator Estimates Your Total Daily Energy Expenditure

This calculator runs the Mifflin-St Jeor equation, a BMR prediction formula derived from 498 adults in the United States in 1990, then multiplies the result by an activity multiplier from the FAO/WHO/UNU 2001 physical activity level (PAL) categories. The error against doubly labelled water, the reference method, sits at ±15-25% for a single person. That is the single most useful fact here. The equation, its derivation population, and the error range define the output and its limits.

The calculator uses the Mifflin-St Jeor equation to estimate basal metabolic rate (BMR). For men: 10 × weight (kg) + 6.25 × height (cm) - 5 × age (y) + 5. For women: 10 × weight (kg) + 6.25 × height (cm) - 5 × age (y) - 161. This equation was published in the American Journal of Clinical Nutrition in 1990. The derivation population was 498 normal-weight, overweight, and obese adults aged 19 to 78 in the United States. It replaced the older Harris-Benedict equation as the preferred prediction formula in many clinical settings because it showed lower systematic bias in modern populations, particularly in people with obesity.

The Mifflin-St Jeor equation was derived from indirect calorimetry measurements on 498 adults. Researchers measured resting metabolic rate (RMR) under controlled conditions and fit a linear regression using weight, height, age, and sex. The result was a prediction equation that estimates total daily energy expenditure only after you multiply it by an activity multiplier. The derivation population is a key limitation: it was a sample of convenience from the United States and has not been validated as accurate for every ethnic group or body composition type.

Population-Specific Bias

The equation overestimates BMR in older adults with sarcopenia and underestimates BMR in very lean people with high fat-free mass. These biases are not small. For a single person, the error against a measured BMR by indirect calorimetry is ±10% in about 80% of non-obese adults. A BMR estimate of 1,600 kcal could land anywhere from 1,440 to 1,760 kcal for four out of five people. For the other one in five, the error is larger.

The Mifflin-St Jeor equation is more accurate than the Harris-Benedict equation for modern populations. The Harris-Benedict equation, published in 1919 and revised in 1984, was derived from a much smaller and less diverse group of European men and women. It systematically overestimates BMR in people with obesity by 5-15% because the derivation population was leaner than the average person today. For a person who is overweight or has obesity, Mifflin-St Jeor gives a lower and more realistic BMR estimate.

When Each Equation Fails

Harris-Benedict overestimates BMR for people with a high body fat percentage. Mifflin-St Jeor does the same for older adults, who lose lean mass with age. Neither equation is precise for a single person. The Katch-McArdle formula uses lean body mass instead of total weight, which makes it more accurate for athletes and bodybuilders. It requires a body fat percentage measurement, and that measurement adds its own error from the method used to estimate it (BIA, skinfolds, or DXA).

The calculator multiplies the Mifflin-St Jeor BMR by an activity multiplier from the FAO/WHO/UNU 2001 physical activity level (PAL) categories. The categories: sedentary (PAL 1.4), lightly active (PAL 1.55), moderately active (PAL 1.7), very active (PAL 1.9), and extra active (PAL 2.2). These PAL categories are not measured values. They are population averages derived from doubly labelled water studies in small groups. For a single person, the error between a TDEE estimated by BMR × PAL and a TDEE measured by doubly labelled water sits at ±15-25%.

Why the Error Is So Wide

The error comes from two sources. First, the BMR equation itself carries ±10-20% error for a single person. Second, the activity multiplier is a coarse category that does not capture non-exercise activity thermogenesis (NEAT) or thermic effect of food (TEF). Two people who select the same activity level on the calculator may have very different actual total daily energy expenditures because one fidgets and paces while the other sits still. The FAO/WHO/UNU 2001 report notes that these PAL categories are designed for population-level energy requirement estimates, not for individual prescription.

BMR prediction equations, including Mifflin-St Jeor, are population-derived. They describe the average BMR of a group of people with your weight, height, age, and sex. They do not describe your BMR. The variation around the prediction is large. A 1990 study by Mifflin et al. reported that the equation predicted BMR within ±10% of measured values for about 80% of the derivation sample. For 20% of the people in that study, the error was larger than ±10%.

What Drives Individual Variation

Lean body mass (fat-free mass) is the single largest determinant of BMR. Two people of the same weight, height, age, and sex can have very different BMRs if one carries more muscle and the other more fat. The Mifflin-St Jeor equation does not account for body composition. It uses total weight, which conflates fat mass and lean mass. An athlete with 15% body fat and a sedentary person with 35% body fat could weigh the same and get the same BMR estimate, but their actual BMRs measured by indirect calorimetry could differ by hundreds of kilocalories per day.

Doubly labelled water is the gold-standard method for measuring total daily energy expenditure (TDEE) in free-living humans. A person drinks water labelled with stable isotopes, and researchers measure the elimination rate over 10-14 days. Studies comparing estimated TDEE from BMR equations multiplied by PAL categories against doubly labelled water measurements consistently show wide errors for individuals. A 2004 FAO/WHO/UNU expert consultation reviewed these comparisons and reported that the error for a single person sits at ±15-25%.

What the Error Means in Practice

An error of ±20% on an estimated TDEE means the true value spans a wide range. The calculator gives you an estimate, not a measurement. It tells you what the population average is for someone with your statistics, not what your actual expenditure is. The only way to get your actual TDEE is to measure it with doubly labelled water. That method is expensive and unavailable outside research settings.

The most common mistake is treating the TDEE estimate as an individually measured number. It is not. The calculator uses a population-derived equation and a coarse activity multiplier. It does not measure your non-exercise activity thermogenesis, your thermic effect of food, your resting metabolic rate, or your body composition. If you use the output to set a calorie target and it is wrong by a large margin because the equation over- or underestimated your BMR, you will not get the result you expected.

Treat the TDEE number as a hypothesis. Track your weight and your intake for two to three weeks and adjust based on the trend. Do not assume the calculator has your number right. It does not.