Why TDEE Calculators Disagree by 300 Calories

Same person, three answers

Take one person: a 30-year-old man, 175 cm, 70 kg, roughly 20% body fat. Run him through the three formulas that power almost every TDEE calculator on the internet and you get three different basal metabolic rates.

Mifflin-St Jeor says 1,649 kcal a day: 10 × 70 + 6.25 × 175 − 5 × 30 + 5. The revised Harris-Benedict equation says 1,696 kcal. Katch-McArdle, which works from lean mass instead of total weight (56 kg lean at 20% body fat), says 1,580 kcal: 370 + 21.6 × 56. Every step of that arithmetic fits on a napkin, so check it if you like.

The spread is 116 kcal. Annoying, but modest. If formula choice were the whole story, TDEE calculators would agree within a bowl of rice. They do not, because the formula is not where the real disagreement lives.

Where each formula comes from

Harris-Benedict is the oldest, published in 1919 from indirect-calorimetry measurements of a few hundred people at the Carnegie Institution. The version calculators use today is usually the 1984 Roza-Shizgal revision, refitted on more subjects. A century of changes in body composition means the original tends to read slightly high for people today.

Mifflin-St Jeor was published in 1990 in the American Journal of Clinical Nutrition, built on about 500 modern adults specifically because Harris-Benedict was drifting. It is the default recommendation of most dietetic bodies, and the default in most calculators. A widely cited 2005 systematic review found it lands within 10% of measured resting energy expenditure for roughly eight in ten non-obese adults — the best hit rate of the common equations, and still wrong by more than 10% for the other two in ten.

Katch-McArdle takes a different input entirely: lean body mass. That makes it the only one of the three that distinguishes a muscular 70 kg from a sedentary 70 kg. The catch is that you rarely know your body-fat percentage precisely. Feed it a guess that is off by five points and you have imported more error than the formula ever removed.

The multiplier is the real culprit

BMR is only the first half of a TDEE calculation. The second half multiplies it by an activity factor: 1.2 for sedentary, 1.375 for light activity, 1.55 for moderate, 1.725 for very active, 1.9 for extremely active. For our 1,649 kcal man, those five choices produce 1,979, 2,267, 2,556, 2,844 and 3,134 kcal.

One step on that scale is about 289 kcal. Read that against the first chart: picking a different formula moves the answer by at most 116 kcal, while picking the adjacent activity label moves it by 289. When two calculators disagree by hundreds of calories, they almost never disagree about your metabolism. They disagree about what "moderately active" means.

And people are bad at this in a consistent direction. A desk job plus three gym sessions a week feels like "moderately active", but the multipliers were defined with all-day movement in mind; that life is closer to 1.375 than 1.55. Choosing the flattering label inflates the estimate by exactly the amount that later gets blamed on a slow metabolism.

What the number can and cannot promise

Even a perfectly chosen equation is a population average applied to one person. Within 10% for eight people in ten also means more than 10% off for two in ten; on a 2,500 kcal budget, 10% is 250 kcal — an entire meal for some. Resting metabolism itself wobbles day to day with sleep, illness, menstrual cycle and ambient temperature, and the same body burns a little less after weight loss than the formulas predict, a phenomenon called adaptive thermogenesis.

For women the equations shift down — Mifflin-St Jeor swaps the +5 for −161, giving 1,483 kcal for the same height, weight and age — but the error bars are the same size. None of this makes the calculators useless. It makes them starting points.

Calibrate against the only data that matters

Your bathroom scale is a better metabolism lab than any equation, because it measures you rather than the average of a 1990 study cohort. The procedure is short. Pick the calculator's maintenance number and eat at it, as consistently as you can log, for two to three weeks. Weigh yourself daily under the same conditions and compare weekly averages, not single days — day-to-day water swings are larger than any real change.

Then read the drift. A kilogram of body fat stores roughly 7,700 kcal, so trending down half a kilo over two weeks means you ate about 275 kcal a day under your true TDEE — nudge the number up by that much. Trending up, nudge down. One cycle of this beats any argument about which formula is superior, because after it, the formula no longer matters: you have replaced the population estimate with your own measured one.

So which one should you pick?

As a first guess: Mifflin-St Jeor, with the activity level one notch below what flatters you. Reach for Katch-McArdle only when your body-fat percentage comes from something better than a mirror estimate — a DEXA scan, or a bioimpedance trend you trust. Treat Harris-Benedict as the historical baseline it is. Then stop optimizing the guess and start the two-week calibration, which quietly absorbs every error the formulas and labels left behind.

The TDEE calculator lets you switch between all three equations on the same inputs, so you can see the 116 kcal spread from the first chart on your own body — and it derives cutting and bulking targets from whichever you choose. If what you are actually wondering is whether your weight itself is in a healthy range, that is a different question and a different tool: the BMI calculator covers it, including where the standard cutoffs mislead.