How to calculate TDEE correctly (and why most online calculators are off by 10%)

Total Daily Energy Expenditure (TDEE) is the calorie number every diet and bulk plan revolves around. The formula has two parts — basal metabolic rate (BMR) and an activity multiplier — and the second one is where most calculators lose accuracy.

The formula

TDEE = BMR × activity_factor

BMR is what your body burns at complete rest. The most accurate predictive formula is Mifflin-St Jeor (1990):

Male: BMR = 10·weight(kg) + 6.25·height(cm) − 5·age + 5
Female: BMR = 10·weight(kg) + 6.25·height(cm) − 5·age − 161

Plug your numbers into our BMR calculator for the BMR value, then multiply by the activity factor below.

The activity factor (where most calculators go wrong)

Standard multipliers:

• 1.2 — Sedentary. Desk job, no exercise. Genuinely sitting most of the day.
• 1.375 — Lightly active. Light exercise 1-3 days per week, or a job with some walking.
• 1.55 — Moderately active. Moderate exercise 3-5 days per week.
• 1.725 — Very active. Hard exercise 6-7 days per week.
• 1.9 — Extremely active. Physical job (construction, professional athlete) plus daily training.

Three common ways people get this wrong:

1. Overestimating activity

Most people walking past a fitness app classify themselves one tier higher than they should. A desk worker who lifts three times a week is “moderately active” — not “very active.” The 25% gap between consecutive tiers translates into hundreds of calories a day.

A sanity check: in a typical 24-hour day, if you spend more than 16 hours sitting or lying down (desk + sleep), your activity tier is at most “lightly active” regardless of what you do for an hour at the gym.

2. Counting one good week as the norm

The activity factor should reflect your typical week, averaged over a month. One hard week followed by three lazy weeks averages to sedentary, not moderately active. Be honest.

3. Ignoring NEAT

Non-exercise activity thermogenesis (NEAT) — fidgeting, standing, walking to the kitchen — varies dramatically between people. Some “sedentary” office workers burn 400 calories more per day than others, purely through unconscious movement. NEAT isn’t in the formula directly; it’s the main reason individuals deviate from their predicted TDEE by ±15%.

The 2-week feedback loop

Any calculator returns an estimate. Your actual TDEE is whatever number, when eaten daily, leaves your weight unchanged over two weeks. The protocol:

1. Get the estimated TDEE from any calculator.
2. Eat exactly that many calories daily for 14 days.
3. Weigh daily, same time, same conditions. Use the 14-day average.
4. If average weight is stable: estimated TDEE was correct.
5. If you gained weight: actual TDEE is lower than estimated. Subtract calories until you stabilise.
6. If you lost weight: actual TDEE is higher. Add calories.

After this calibration, the activity factor that matched your real outcome is your honest factor — use that one going forward, not the one your gym app picked.

When BMR estimates themselves are off

Mifflin-St Jeor predicts BMR within about ±10% for most adults. Where it’s less accurate:

• Highly muscular people. Predicts low. Muscle is more metabolically active than the formula assumes.
• Obese individuals. Predicts high. Adipose tissue burns fewer calories per kilogram than lean mass; Mifflin-St Jeor doesn’t distinguish.
• People with hyperthyroid or hypothyroid conditions. Thyroid drives basal metabolism by ±15-30%; the formula doesn’t know.

For most general-population users, Mifflin-St Jeor is the right starting point. The 2-week feedback loop catches the residual error.

Worked example

A 34-year-old woman, 168 cm, 68 kg, desk job with three 45-minute strength sessions per week.

• BMR (Mifflin-St Jeor): 10·68 + 6.25·168 − 5·34 − 161 = 680 + 1050 − 170 − 161 = 1399 kcal.
• Honest activity tier:three lifts a week on a desk-job baseline is “lightly active” (1.375), not “moderately active” (1.55). The gym hours are a small fraction of the week; the desk hours dominate.
• Estimated TDEE: 1399 × 1.375 ≈ 1924 kcal/day.
• 14-day calibration: eats 1925 kcal daily, loses 0.4 kg over the two weeks. 0.4 kg ≈ 3080 kcal deficit ÷ 14 days ≈ 220 kcal/day. Actual maintenance is closer to 2145 kcal. The activity factor that matches her body is 1.53 (between “lightly” and “moderately active”) — exactly the NEAT-driven spread the formula can’t see.

Common mistakes

• Trusting the wearable’s “calories burned” number. Wrist-based heart-rate devices over-estimate exercise burn by 30-90% in peer-reviewed comparisons (Shcherbina et al., J Pers Med, 2017). Don’t add “exercise calories back” using a Fitbit/Apple Watch reading on top of a TDEE that already includes an activity multiplier — you double-count.
• Recalculating after every weight change.BMR depends on lean mass, not total weight; a 2 kg fat-loss drop barely shifts BMR. Recalculate when you’ve changed body composition meaningfully (≥5 kg) or every 12 weeks, whichever comes first.
• Picking the highest of multiple calculator outputs. Different formulas (Harris-Benedict, Katch-McArdle, Cunningham) produce a 200-300 kcal spread. Anchoring on the highest number is motivated reasoning. Pick one validated equation (Mifflin-St Jeor for the general population, Katch-McArdle if you have a measured lean-mass value) and stick with it.
• Ignoring weekend creep. Eating at maintenance Mon-Fri and adding 800 kcal on Sat-Sun is a ~230 kcal/day weekly average surplus. The calibration loop must include weekends to be honest.
• Cutting too aggressively for too long.Sustained deficits >25% trigger measurable adaptive thermogenesis (Fothergill 2016): metabolic rate drops below the predicted level and stays low for years. Modest deficits (15-20%) preserve more long-run metabolic flexibility.

Edge cases

• Over-60s. Pontzer et al. (Science, 2021) showed that TDEE per kg fat-free mass stays stable from age 20 to 60, then drops ~0.7% per year. Sedentary predictions for ages 60+ over-estimate by 10-15%; lean-mass loss compounds the gap.
• Pregnancy and lactation.Add ~340 kcal/day in the second trimester, ~450 in the third, and ~500 during exclusive breastfeeding (USDA DRI). The Mifflin-St Jeor formula doesn’t handle either case — defer to your OB or RD.
• Recent GLP-1 medications (semaglutide, tirzepatide). Appetite suppression makes calorie intake hard to track and lean-mass loss can run 25-40% of total loss without resistance training. Calorie targets should be set under clinical supervision.
• Type 1 diabetes.Insulin dosing interacts with carbohydrate intake in ways the formula doesn’t model. Calorie planning needs an endocrinologist or certified diabetes educator.

What “maintenance calories” actually means

Your maintenance calorie level is the daily intake that produces zero net weight change over time. It’s synonymous with TDEE only if your body composition is stable. During a deliberate cut or bulk, daily intake is by definition different from maintenance — that’s the whole point.

Common targets relative to maintenance:

• Aggressive cut: maintenance − 25-30%.
• Moderate cut: maintenance − 15-20%.
• Lean bulk: maintenance + 10-15%.
• Recomp (build muscle while losing fat): maintenance ± 0-5%, depends on training stimulus and protein intake.

Get your maintenance number from our BMR & TDEE calculator, then verify it with the 2-week loop, then adjust from there. For the BMI side of the picture, see BMI vs body fat percentage.