Why Your TDEE Is Probably 500 Calories Too High (What 29 Measured Users Showed Us)

A 9-minute read on the gap between the number a calculator gives you and the number your body is actually running on — measured, not argued.

---

TL;DR

Every TDEE calculator, ours included, builds your daily burn the same way: a resting-metabolism formula, plus an allowance for how active you are. The resting part is fine. The activity part is roughly twice too generous for the people who actually use these tools.

We know because we checked. Twenty-nine NanoRhino users logged their meals and stepped on a scale often enough over seven weeks that we could work out, from their own numbers, what their bodies were really burning. Then we looked up what our calculator had told each of them on the day they signed up.

• Real burn ran a median 511 kcal/day below the sign-up number. Twenty-two of the 24 people with a sign-up number came in below it.
• Real burn was 1.16 × resting metabolism. The calculator had assumed 1.38 ×.
• Everything the formula stacks on top of resting metabolism — job, steps, chores, workouts, digestion — added up to about 600 kcal/day on paper and 300 in reality.
• Had these people simply eaten what the calculator's plan told them to, 14 of the 29 would have stalled or gained weight, while being promised a loss of half a pound a week.

So we changed the calculator. As of today, the number you get from the NanoRhino TDEE calculator is anchored to what measured users actually burn, not to the textbook stack — with the textbook figure shown alongside so you can see the difference. The rest of this post explains where the 500 calories go.

---

How every calculator gets to a number

There are two layers.

Layer one: resting metabolism (BMR). What you burn lying still. The standard tool is the Mifflin–St Jeor equation from 1990, and it is genuinely good: in the largest validation review, it landed within 10% of measured resting metabolism for about 82% of non-obese adults and about 70% of obese adults — better than any competing equation [1]. This layer is not the problem.

Layer two: everything else. Your job, your steps, your commute, the cooking and the kids, your workouts, and the energy it takes to digest food. Traditional calculators compress all of that into one dropdown — sedentary 1.2, lightly active 1.375, moderately active 1.55, active 1.725, very active 1.9 — and multiply. Component calculators like ours try to do better by asking about each piece and adding them up.

Either way, layer two is where the number goes wrong, for four reasons that compound.

Reason 1: people overestimate their own activity, a lot

The classic study here is thirty years old and still uncomfortable. Researchers took people who insisted they could not lose weight on a low-calorie diet, and measured them with the gold-standard method (doubly labelled water). On average they were under-reporting intake by 47% and over-reporting physical activity by 51% — and were entirely sincere about both [2].

That is not a character flaw; it is how memory works. "I'm on my feet all day" is a feeling about how tired you are, not a measurement of how many calories you spent. Any calculator that takes the self-description at face value inherits the inflation.

Reason 2: extra activity doesn't add on top the way the arithmetic assumes

The activity table treats your burn as resting metabolism plus a fixed bonus for each thing you do. Bodies don't cooperate. Herman Pontzer's group measured total energy expenditure across a wide range of activity levels and found it plateaus: beyond a moderate level, more movement barely raises total burn, because the body quietly economises elsewhere [3]. The "very active 1.9 ×" bucket describes a number that most people's physiology never reaches.

Workouts have the same problem in miniature. A treadmill says 400 calories; it is counting the calories you would have burned sitting on the couch during that hour too, and it isn't counting the afternoon you spend less fidgety because you trained. The honest net figure is usually well under half of the display.

Reason 3: the formulas were validated on a different population

The activity multipliers trace back to work with younger, leaner, more physically active reference groups. Look at who actually uses a weight-loss calculator: our measured 29 were mostly women, median age in the mid-50s, with a median BMI in the mid-30s. That is exactly the population where even the resting equation is least accurate [1], and where the activity assumptions are most wrong — a desk job, a car, an hour of TV, and a genuine 30-minute walk five days a week is sedentary by the arithmetic, and the walk buys about 100 calories, not 300.

Reason 4: the number you'll live with is measured in "as-logged" calories

This one is subtle and it matters. When you follow a calorie target, you follow it through a food log — and every food log misses things: the oil, the tasting spoon, the second handful. Studies of logging consistently find shortfalls in the 10–30% range, more for people with more to lose [2].

That means the useful maintenance number — the one where "eat this many logged calories" holds your weight steady — is lower than your physiological burn, by roughly your logging gap. Our measurement is taken in those same as-logged units, which is exactly why it is the right anchor for a target you will track against, and why it comes in lower than a lab would. A calculator that reports pure physiology hands you a number you can only hit by logging perfectly. Nobody logs perfectly.

What the measurements actually looked like

The method is simple and old: over a stretch of weeks, put what someone ate next to what their weight did. If the scale held still while the log said 1,500 a day, the body was burning about 1,500 a day in those units. If it dropped a pound a week, the body was burning about 500 more than the log. Do this over enough weeks and enough weigh-ins and the noise averages out.

Two of the 29, anonymised:

• A 72-year-old retiree, 5'4", about 188 lb. The calculator said 1,860. She ate 1,340 a day — carefully, with photos — and her weight did not move for three weeks. Her measured burn: about 1,330. At 1,340 she was, correctly, maintaining. The formula had promised her half a pound a week of loss that could not happen.
• A 43-year-old mother of two, 5'4", about 240 lb. The calculator said 2,710. She ate to her plan, around 2,000–2,100 a day, and sat at the same weight for three weeks. Her measured burn: about 2,120. Her real deficit was a hundred calories, not six hundred.

Neither of them was doing anything wrong. They were doing exactly what a wrong number told them to.

Across the group: the calculator's sign-up figure was 1.38 × resting metabolism; the measured figure was 1.16 ×. The middle half of people fell between 1.02 × and 1.28 ×. Five of the 29 burned more than the formula said — mostly younger and heavier people with genuinely physical days — so this is a bias, not a law, and it cuts both ways.

What eating to the wrong number does

We ran the counterfactual on all 29: suppose each person had eaten exactly the calorie target their original plan gave them.

• The plan promised a median loss of 0.56 lb per week.
• Their measured burn says they would actually have lost 0.27 lb per week.
• 12 would have gained weight. 2 would have stalled. The rest would have lost, some faster than promised.

That gap — about 300 calories a day of "permission" the formula hands out — is the mechanism behind the single most common complaint in weight loss: I'm doing everything right and nothing is happening. Usually nothing is broken. The starting number was.

What we changed in the calculator

We didn't throw the model out. Resting metabolism stays exactly as it was; it's the well-validated part. The change is in the layer on top:

• The low end of your range now keeps only half of the textbook's above-resting estimate (job, steps, chores, workouts, digestion), and never drops below 1.12 × your resting metabolism.
• The high end keeps 80% of it, so the range still reaches the people who really do burn more.
• The calculator's default recommendation is the low end — it always was — so the number most people act on moved down by roughly the amount we measured.
• The textbook figure is still shown next to the calibrated one. If you'd rather trust the formula, the information is there.

What that looks like for four typical profiles:

Profile	Textbook range	Calibrated range	Recommended
Woman, 45, 5'4", 209 lb, desk job, ~4,000 steps, no workouts	1,870–2,060	1,770–1,960	1,770
Woman, 62, 5'4", 191 lb, desk job, 30-min walk × 5/week	1,790–2,000	1,600–1,880	1,600
Man, 46, 6'4", 265 lb, on-your-feet job, ~8,000 steps, lifts × 3/week	3,180–3,450	2,680–3,200	2,680
Man, 32, 5'10", 176 lb, desk job, ~6,000 steps	2,140–2,260	1,970–2,160	1,970

Against the 24 people with a sign-up number, the calibrated version cuts the median error from 511 to about 175 kcal/day and turns a 19-to-1 over-estimate into a roughly even split. We deliberately left a small lean toward the high side: telling someone to eat too little is the mistake that makes people hungry and quit, and the coach that follows sign-up corrects the number from your own weigh-ins and meals anyway.

We'll re-fit the constants when the measured group is a couple of times larger. Twenty-nine is enough to be sure of the direction and the rough size; it is not enough to be precise about a 37-year-old nurse versus a 61-year-old accountant.

If you use any calculator, ours or anyone's

1. Treat the number as a ceiling, not a fact. Take the low end of whatever range you're given.
2. If you're over 50, have a desk job, or have more than 50 lb to lose, mentally subtract 10–20% from any activity-multiplier result before you subtract a deficit.
3. Don't eat back exercise calories. The display on the machine is gross, not net, and your body will spend less elsewhere.
4. Measure yourself. Two weeks of honest logging and a scale reading every few days tells you more about your real burn than any form. If the scale doesn't move on a target the calculator called a 500-calorie deficit, the target was wrong, not you.
5. Never go below your floor. A calibrated number that lands near your resting metabolism means calories are not your lever — movement is. No calculator should send you under your minimum, and ours will not.

The formula's job is to get you through week one. After that, the best information about you is you.

---

Try it

The recalibrated NanoRhino TDEE calculator is free and asks no email. If you'd like the number to keep correcting itself from your own weigh-ins and meals, start messaging your coach — that's what the deficit loop is for. Pricing is $10 per pound you actually lose, capped at $500 — no subscription; if the number is wrong for you, it doesn't cost you.

---

Sources

1. Frankenfield D, Roth-Yousey L, Compher C. "Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review." Journal of the American Dietetic Association, 2005.
2. Lichtman SW et al. "Discrepancy between self-reported and actual caloric intake and exercise in obese subjects." New England Journal of Medicine, 1992.
3. Pontzer H et al. "Constrained total energy expenditure and metabolic adaptation to physical activity in adult humans." Current Biology, 2016.
4. Mifflin MD et al. "A new predictive equation for resting energy expenditure in healthy individuals." American Journal of Clinical Nutrition, 1990.

---

Written by Xiaomeng Li, Co-founder & CTO of NanoRhino. I'm an AI engineer, not a clinician — no medical or dietetic credential, and nothing here is medical advice. The measurements above come from a small, self-selected group of our own users with no control group; they show a direction and a rough size, not a clinical result. If you have a medical condition, take medication that affects appetite or metabolism (including GLP-1s), are pregnant, or have any history of disordered eating, talk to a doctor or a registered dietitian before changing what you eat. Our tools enforce per-person minimum intake floors and will not set a target below them.

Built with care by Link Heart Limited in Houston, Texas.