What are BMR, NEAT, TEF and EAT?

The four components of daily energy expenditure. BMR is what keeps you alive at rest, typically 60 to 70 percent of the total. NEAT is non-exercise activity — walking, standing, fidgeting. TEF is the energy cost of digesting food, around 10 percent. EAT is deliberate exercise, usually 5 to 15 percent and smaller than people expect.

Why split TDEE into four numbers?

Because they behave completely differently and the single figure hides the two you can actually influence. BMR barely moves day to day. NEAT varies by up to around 2,000 calories a day between people and is the component that quietly falls during a long deficit. Seeing them separately is the difference between my metabolism broke and I stopped moving as much.

How is the thermic effect of food calculated?

From the macros you actually logged, rather than as a flat percentage. Protein has a substantially higher thermic effect than carbohydrate or fat, so a high-protein day genuinely costs more to digest. That is why the number moves with what you ate rather than only with how much.

Why is my exercise share smaller than I expected?

Because it usually is. An hour of hard training might cost 500 to 800 calories against a total of 2,800 — meaningful, but a fraction of the day. This is the single most common misjudgement in energy balance, and seeing the split is the fastest way to correct it.

Does this screen replace my adaptive maintenance number?

No. Adaptive maintenance is measured from your trend weight against your intake and is the more reliable figure. This breakdown is a decomposition using published models, and it exists to explain the shape of your expenditure rather than to replace the measurement.

See Where Your Calories Actually Go

The four components

Component	Typical share	Behaviour
BMR	60–70%	Barely moves day to day
NEAT	10–30%, highly variable	Falls quietly during a long deficit
TEF	~10%	Scales with intake, especially protein
EAT	5–15%	Smaller than people expect

The day's burn split four ways. Each slice opens its own explanation, because the four are different kinds of number with different uncertainty attached to each.

BMR

What keeps you alive at rest. Estimated with Mifflin-St Jeor:

men:   10 × kg + 6.25 × cm − 5 × age + 5
women: 10 × kg + 6.25 × cm − 5 × age − 161

Chosen over Harris-Benedict because it is more accurate in the general population, and over Katch-McArdle because that needs a body-fat figure Base would have to estimate — importing that estimate's error into the largest component of the total.

It carries a standard error of a couple of hundred calories. That is not a flaw in the implementation; it is what prediction equations do, and it is one reason measured maintenance is the more trustworthy number.

NEAT

Walking, standing, fidgeting, everything that is movement but not training.

It is the most interesting component, because it varies by up to around 2,000 calories a day between people, and because it is the one that quietly falls during a sustained deficit. You move less without deciding to: fewer steps, less standing, less fidgeting.

That fall is a large part of why a deficit stops working at the same intake, and it is invisible in any app that reports one TDEE number.

TEF

The cost of digesting food, computed from the macros you logged rather than as a flat percentage.

Protein has a substantially higher thermic effect than carbohydrate or fat. A 200 g protein day genuinely costs more to process than a 100 g one, and computing it from what you actually ate is what makes the number move for the right reason.

EAT

Deliberate exercise, from your health store's own workout model.

This is where people are most consistently wrong. An hour of hard training might cost 500–800 calories against a daily total of 2,800. Meaningful, but a fraction — and far less than the intuition that a hard session "earns" a large amount of food.

Why this matters for a hybrid athlete specifically

The four-way split is most useful at the edges of a training block, and the case it was built for is coming off a large endurance load.

When Ironman-scale volume stops, two things happen at once: EAT collapses, and NEAT is often suppressed as well because you are tired. Measured TDEE drops sharply, and a single number offers no way to tell whether that was the training stopping or something metabolic.

The split answers it immediately. Exercise going from 900 to 100 with NEAT flat is training volume. Exercise flat with NEAT down 400 is a behavioural change worth noticing.

This is a model, not a measurement

Important framing. This screen decomposes expenditure using published models. It is not the same thing as adaptive maintenance, which is back-calculated from your own scale and intake.

When they disagree, trust the measurement. The decomposition exists to explain the shape of your expenditure, not to replace the figure derived from your own data.

Each slice opens its own explanation rather than sharing one, because the four are genuinely different kinds of number: a prediction equation with a known error, an estimate of incidental movement, an arithmetic result from logged macros, and a platform's proprietary workout model. Sending all four to one sheet would have to describe all four at once, and the uncertainty on each is exactly what is being asked about.

Common mistakes

Reading the exercise share and concluding training does not matter. It matters for adaptation. Its direct calorie cost is simply smaller than intuition says.

Treating BMR as precise. Couple-hundred-calorie standard error.

Assuming NEAT is fixed. It is the most variable component and the most responsive to a deficit.

Using this instead of measured maintenance. The measurement wins.

Expecting the four to match your health app exactly. Your platform reports resting and active; this splits those differently and adds TEF, which no fitness app reports.

Ignoring TEF on a high-protein bulk. At 200 g of protein it is not a rounding error.

What to do next

Open the breakdown after a rest day and after a hard training day and compare the exercise slice. The difference is what a session genuinely costs you — and it is usually smaller, and therefore easier to eat back, than expected.