How to calculate your TDEE, and what it can't say

TDEE — total daily energy expenditure — is the number every calorie target is built from. It is worth understanding what it contains and why every calculated version of it is a hypothesis rather than a fact.

The four components

Basal metabolic rate, 60 to 70 per cent of the total. The cost of being alive at rest, driven mostly by body mass and how much of it is lean tissue. See BMR.

Thermic effect of food, roughly 10 per cent of intake. The energy spent digesting and storing what you eat. Protein costs the most to process.

Exercise activity, often under 10 per cent for most people. An hour of training is real and it is not the bulk of the day.

Non-exercise activity, the wildcard. Walking, standing, stairs, fidgeting, carrying things. This varies by several hundred calories a day between two people of the same size, which is why it defeats prediction — steps, NEAT, and the calories you burn standing up.

The calculator method

Estimate BMR, then multiply.

The Mifflin–St Jeor equation is the usual choice:

• Men: (10 × weight in kg) + (6.25 × height in cm) − (5 × age) + 5
• Women: (10 × weight in kg) + (6.25 × height in cm) − (5 × age) − 161

Then multiply by an activity factor:

• Sedentary, desk job, little movement: 1.2
• Light, some walking or one or two sessions a week: 1.375
• Moderate, three to five sessions plus reasonable daily movement: 1.55
• Active, six sessions or a physical job: 1.725
• Very active, physical job plus daily training: 1.9

A 70 kg, 175 cm, 30-year-old man: BMR ≈ 1,649. At 1.55, TDEE ≈ 2,556.

Where this goes wrong

The multiplier is the weak link. Most people choose one too high. A desk job with three gym sessions is generally 1.4 to 1.5, not 1.7, because three hours of training a week is a small fraction of 168 hours.

Individual variation is large. Two people matching on every input can differ by 300 kcal or more in real expenditure.

Body composition is invisible to the equation. Lean tissue costs more to maintain than fat tissue, and weight alone does not distinguish them.

So treat the output as a starting hypothesis, choose the lower multiplier if unsure, and expect to correct it.

The measurement method, which is better

Log everything for two to three weeks without changing your eating. Weigh most mornings and compare weekly averages.

Flat weight means your average intake is your TDEE. Falling by 0.3 kg a week means your TDEE is roughly your intake plus 330. Rising by the same means it is your intake minus 330.

This measures the thing directly. It takes three weeks and it removes every source of error in the equation at once. The procedure is in maintenance calories, and how to find yours.

What TDEE cannot tell you

It is not fixed. It moves with season, illness, sleep, stress and how much you happen to be walking this month. Treat it as a band of a couple of hundred calories rather than a point.

It falls as you lose weight. Partly because a smaller body costs less, partly because incidental movement declines. This is why targets need recalculating every four to six weeks, as set out in how to calculate your calorie deficit properly.

It says nothing about composition. Two people at the same TDEE eating the same deficit can lose very different proportions of fat and muscle, depending on protein and training — losing fat while keeping the muscle you built.

Do not eat back exercise calories on top

A common error: calculate TDEE with an activity multiplier that already includes your training, then add the calories your watch reports for each session.

That double-counts the exercise, and the watch's figure is unreliable in its own right — are the calories-burned numbers worth anything.

Pick one approach: either a multiplier that includes your usual training, or a sedentary baseline plus logged exercise. Not both.

The practical sequence

Use a calculator to get a starting number. Choose the lower activity multiplier. Eat at it for two to three weeks while logging honestly. Compare your weight trend with what it predicted. Adjust.

After that loop you have a figure that came from your own body rather than from a regression fitted to other people, and every subsequent decision is built on something real.