TDEE math, made honest

Your total daily energy expenditure is the anchor for everything this protocol does, and almost every way of estimating it is wrong by 10–20% — sometimes more. This page walks what the calculators actually compute, where the error sneaks in (it is almost always the activity factor, not the formula), and how to replace a five-minute online guess with a two-week empirical measurement that beats any calculator on earth.

The Number That Anchors Everything

Total daily energy expenditure (TDEE) is the sum of everything your body burns in a day: resting metabolism, digestion, deliberate exercise, and all the fidgeting, walking, cleaning, and standing you never think about. It matters because the entire deficit protocol rests on it — eating below this number is the whole game. If the estimate is high by 300 kcal, a "500-calorie deficit" quietly becomes 200, and the scale stalls for weeks while you assume you are doing something wrong. If it is low by 300 kcal, you are cutting harder than intended and paying in muscle and mood.

The uncomfortable fact is that no calculator knows your number, and no equation can tell you the truth from a height, weight, and age. A properly fitted prediction — using the best-validated formula — still lands within roughly 10% of true resting expenditure for about two-thirds of people, which means for the other third it is off by more. Add the activity factor on top, where the real guesswork lives, and the honest error band on an online result is closer to 20%. That is not a reason to ignore the math; it is a reason to treat the calculator as a starting point and install a measurement step.

Where the Burn Actually Goes

Before worrying about formulas, see the shape of a typical day. For a moderately active adult, resting metabolism consumes the large majority of the total — around 60–70% — and is largely a function of body size, muscle mass, and age. Diet-induced thermogenesis, the cost of digesting what you ate, is a modest single-digit slice. Structured exercise is often smaller than people expect — an hour of moderate training might account for a few hundred calories. And non-exercise activity, the hidden category, ranges from near-zero in a sedentary person to a thousand-plus calories a day in someone who walks, stands, and moves constantly.

The useful lesson is about priorities. Because resting metabolism is the biggest line, people assume they should fixate on it — but it is also the hardest line to move meaningfully, and chasing a higher BMR via weight-loss tricks is mostly theater. The two lines that actually vary, and vary enormously, are non-exercise activity and structured exercise, and those are exactly the ones the calculators cannot see.

The Formula Is the Reliable Part

The best-known predictive equation is named for its developers, Mifflin and St Jeor, and was published in 1990 — a regression fit to measured resting energy expenditure in hundreds of healthy adults (Mifflin et al., American Journal of Clinical Nutrition, 1990). It improves on older formulas (Harris–Benedict dates to 1919) and has become the reference standard for everyday use. It takes sex, height, weight, and age, returns an estimate of resting metabolism, and that estimate is then multiplied by an activity factor chosen from a dropdown to give your printed TDEE.

Here is the honest hierarchy of trust. The formula itself has been validated repeatedly and is a reasonable group-level tool. The inputs are usually accurate — except when people enter "goal weight" instead of current weight. The activity factor is the weak link: it is one of five prepackaged numbers (sedentary, light, moderate, very active, extra active), each covering a huge range, and nobody can reliably tell which bucket they are in. Most people, asked to classify themselves, pick "moderately active" — and most who pick it are overestimating how much they move.

The Activity-Factor Lie

The activity multiplier is where calculators quietly go wrong, and it fails in a predictable direction. The official multipliers run from about 1.2 (sedentary) up to 1.9 (very active), a swing of hundreds of daily calories at common body weights. Research using devices that count actual movement consistently finds people overestimate how active they are, sometimes by a wide margin. The practical translation: pick the sedentary or light bucket unless you have concrete evidence you move more, and treat even that choice as a hypothesis, not a fact.

The Two-Week Empirical Test

The reliable way to find your real TDEE is not a better formula — it is a short, boring measurement: eat at an estimated maintenance level for two weeks, weigh yourself daily under the same conditions, and average the results. Your true maintenance is the intake at which the trend line is flat. The default posture toward the scale is covered in depth on the sibling page on scale noise; here is the test itself.

• ⚖️ Get an honest starting number. Use the Mifflin-St Jeor equation at your current weight with a sedentary-to-light multiplier (1.2–1.35). Write it down; this is your hypothesis, not your answer.
• 🥄 Eat at that number precisely for two weeks. This is the one period where food weighing matters most — the calibration page for weighing, the point being you need intake accuracy here for the output to mean anything.
• 📉 Weigh every morning after the bathroom, before food or drink. Same conditions, same scale, seven days of data per week.
• 🧮 Compute the seven-day averages. Compare week one to week two. Flat average (±0.3 kg) means you found maintenance. Dropping means your real TDEE is lower — the estimate was high. Rising means the reverse.
• 🔁 Adjust in 150–200 kcal steps. If the average crept up, subtract a step and repeat for another week. Two weeks is usually enough to land within a useful range.
• 📝 Record the winner. That number — not the app's — is your reference for the deficit, and it stays valid until your weight, activity, or body size changes materially.

The Recheck Rule

TDEE is not a fixed constant; it is a moving target. As you lose weight, resting metabolism drifts down with the mass — roughly a few kcal per kilogram lost — and the body also defends against a deficit by trimming non-exercise activity without you noticing. The maintenance page for this series covers the long arc; the practical rule here is to re-run the two-week test (or a lighter variant) after every 4–5 kg of loss, and whenever the scale stops matching the deficit you think you are running. You are measuring a living process, not carving a number in stone.

Practical Rules for the Estimate

Questions, Answered Briefly

• 📉 Three different calculators gave me three different numbers. Normal and expected — the formulas are similar, but activity factors and rounding differ. Pick one tool, use it consistently, and treat its output as a starting point for the two-week test, not as three competing verdicts.
• 🏃 I exercise five days a week — do I count as "very active"? Structured exercise rarely justifies the top multiplier by itself. The category is about total daily movement, not just workouts. Unless you walk or stand a great deal on top of training, 1.35–1.55 is a more honest band.
• 🧬 Do devices with heart rate give the real number? They are directionally useful and better than pure self-report, but still carry meaningful error — treat them as evidence you move more, not as ground truth. The scale test remains the arbiter.
• 🩺 I'm taking medication that affects weight. Does this still apply? Some medications shift appetite or resting metabolism, so your TDEE may sit outside the predicted range — another reason the empirical test matters more. Run changes by your clinician; nothing here prescribes.
• ⏳ Do I really need to weigh food for a whole two weeks? It is the price of a number you will lean on for months. If two weeks feels impossible, do one week and accept a wider error bar — and know the deficit built on it inherits that uncertainty.

Related Topics