How to estimate your maintenance calories (and why the number is only a starting point)

What "maintenance calories" actually means

Maintenance calories are the intake that keeps your weight roughly stable over time: eat above it and you gain, eat below it and you lose. It isn't a fixed number — it moves with your weight, activity level, sleep, and even the season — and no formula can measure it directly without a metabolic chamber. What every calculator (including the ones on this site) actually does is estimate your basal metabolic rate (BMR), the energy your body burns at complete rest, and then scale that up for activity to get total daily energy expenditure (TDEE), which is the practical stand-in for "maintenance calories."

Three formulas, three different starting numbers

The Mifflin-St Jeor equation (1990) is: 10 × weight(kg) + 6.25 × height(cm) − 5 × age + 5 for men, or −161 for women. It was derived from indirect calorimetry measurements — the gold-standard way to measure resting energy use — on 498 healthy adults, published in the American Journal of Clinical Nutrition by Mifflin and colleagues (PubMed).

The Harris-Benedict equation is older: first published in 1919, then revised by Roza and Shizgal in 1984 using updated coefficients (PubMed). It uses the same four inputs — sex, age, height, weight — but different constants, and it consistently runs a little higher than Mifflin-St Jeor for the same person.

The Katch-McArdle formula takes a different input entirely: BMR = 370 + 21.6 × lean body mass (kg). It comes from the exercise-physiology textbooks of Katch, McArdle and Katch rather than a single study, and it needs a body-fat percentage — from the body fat calculator, a caliper reading, or a DEXA scan — because it's driven by how much metabolically active lean tissue you carry, not your total weight.

Worked example: one person, three numbers

Take a 30-year-old man, 180 cm tall, 82 kg, at roughly 18% body fat (67.2 kg of lean mass):

• Mifflin-St Jeor: 10 × 82 + 6.25 × 180 − 5 × 30 + 5 = 820 + 1,125 − 150 + 5 = 1,800 kcal
• Harris-Benedict (revised): 88.362 + 13.397 × 82 + 4.799 × 180 − 5.677 × 30 ≈ 1,880 kcal
• Katch-McArdle: 370 + 21.6 × 67.2 ≈ 1,822 kcal

That's roughly an 80 kcal spread across formulas for the same man — small day to day, but enough to matter over a few months of tracking. A leaner or more muscular person would see Katch-McArdle diverge further from the other two, since it's the only formula of the three that responds to body composition rather than inferring it from height and weight.

Which formula to trust

A systematic review by Frankenfield and colleagues, published in the Journal of the American Dietetic Association (2005), compared several resting-metabolic-rate prediction equations against directly measured values and found Mifflin-St Jeor predicted actual RMR within 10% more consistently than Harris-Benedict, in both non-obese and obese adults (PubMed). That's why Mifflin-St Jeor is the default formula in the BMR and TDEE calculators on this site. Katch-McArdle is worth using instead if you know your body-fat percentage and you're notably leaner or more muscular than average, since two people of identical height and weight can have genuinely different BMRs depending on how much of that weight is lean tissue.

From BMR to a daily number

BMR alone isn't what you eat to maintain weight — it's what your body burns lying still. To get TDEE, BMR is multiplied by an activity factor: roughly 1.2 for a sedentary desk job, up to 1.9 for very hard daily training plus a physical job. For the 1,800 kcal Mifflin-St Jeor result above, a moderately active lifestyle (factor 1.55) gives a TDEE around 2,790 kcal — the number that actually answers "how much can I eat to stay the same weight."

Why the number is only a starting point

Individual metabolic rates vary by roughly ±10–15% around any formula's prediction, even the best-validated ones, because genetics, thyroid function, medications and NEAT (the calories burned in unconscious fidgeting and daily movement) aren't captured by sex, age, height and weight alone. The only way to convert an estimate into a number you can trust for your own body is to test it: eat at the calculated target for two to three weeks, weigh yourself under consistent conditions, and compare your actual weight trend to what the formula predicted. If you're losing or gaining faster or slower than expected, adjust intake by 100–200 kcal and reassess — your own weight trend is a more accurate measurement than any equation.

---

This article explains general formulas used in fitness and nutrition research. It is not medical advice and doesn't account for your individual health, medications or conditions — talk to a doctor or registered dietitian before making significant changes to your diet.