Why the Deficit Shrinks as You Do
Metabolic adaptation explained: the arithmetic of a lighter body plus adaptive thermogenesis, with a worked weekly example and the uncertainty in each figure.
A calorie deficit is not a fixed quantity. It is the gap between two numbers that both move, and the one that moves faster is expenditure. Maintenance energy expenditure falls as body mass falls, so a diet that starts at a 500 kcal daily deficit does not stay at 500 kcal. The arithmetic of a smaller body accounts for most of the decline. A further reduction, adaptive thermogenesis, sits on top of it and is real, measurable, and smaller than the framing that circulates online.
Two effects, usually conflated
The first effect is arithmetic. A body of lower mass costs less to run at rest and less to move. The second effect is adaptive thermogenesis: a reduction in energy expenditure beyond what the change in mass and body composition predict. The first is large and predictable. The second is smaller, variable, and partly reversible.
The distinction matters because the two are routinely merged into a single claim that the metabolism is damaged. It is not damaged. It is responding to a smaller body and to a sustained energy deficit, and the response is a change in rate, not a change in kind.
The arithmetic effect, quantified
Resting metabolic rate (RMR) is estimated from body mass, height, age, and sex. The Mifflin-St Jeor equation is the common clinical choice:
- Men: RMR = 10 × weight(kg) + 6.25 × height(cm) − 5 × age(y) + 5
- Women: RMR = 10 × weight(kg) + 6.25 × height(cm) − 5 × age(y) − 161
Total daily energy expenditure (TDEE) is RMR multiplied by an activity factor, typically 1.2 to 1.9. The Mifflin-St Jeor estimate carries a standard error of roughly 10% in adults, meaning the true value for an individual commonly falls within about ±10% of the calculated figure. That band is not a rounding detail; it is the width of the target.
Because RMR depends on weight, a loss of mass lowers RMR directly. A loss of 10 kg lowers the weight term by 100 kcal per day in the equation, before any activity change. That is the arithmetic effect, and it is the larger share of the decline.
Adaptive thermogenesis, quantified
Adaptive thermogenesis is the component of the decline not explained by mass and composition. Estimates from controlled feeding studies place it at roughly 50 to 150 kcal per day below predicted expenditure during active weight loss, with wider individual variation. It is not a fixed penalty and it is not permanent in the way the phrase “damaged metabolism” implies. It attenuates with energy availability and tends to reverse, partly or fully, when energy intake is restored to maintenance.
The magnitude is contested. Some studies report larger values in specific conditions, such as very low energy availability or high prior weight loss. The honest summary is a range, not a point: a further reduction beyond mass-predicted expenditure, commonly in the low hundreds of kilocalories per day, not thousands.
A worked example, week by week
Start with a 35-year-old woman, 165 cm, 80 kg, moderately active (activity factor 1.5). The calculation is shown step by step so it can be repeated with other numbers.
Step 1: RMR = 10 × 80 + 6.25 × 165 − 5 × 35 − 161
Step 2: RMR = 800 + 1031.25 − 175 − 161
Step 3: RMR = 1495 kcal/day
Step 4: TDEE = RMR × 1.5 = 1495 × 1.5 = 2243 kcal/day
Step 5: 10% band on TDEE = 224 kcal, so TDEE is roughly 2020–2470 kcal/day
Step 6: Intake set at 1743 kcal/day gives a nominal deficit of 500 kcal/day
Step 7: Predicted loss ≈ 500 / 7700 = 0.065 kg/day, or about 0.45 kg/week
The 7,700 kcal per kilogram figure describes the energy density of adipose tissue, not of body weight, and it is wrong in two directions: early loss includes water and glycogen, which overstates fat loss, and later loss is slowed by falling maintenance, which understates the intake needed to keep losing at the same rate.
Now carry the same person forward. Assume 0.45 kg/week for the first two weeks, then recompute RMR at the new weight each week, holding intake at 1743 kcal/day and the activity factor at 1.5.
| Week | Weight (kg) | RMR (kcal/day) | TDEE (kcal/day) | Intake (kcal/day) | Deficit (kcal/day) | Predicted loss (kg/week) |
|---|---|---|---|---|---|---|
| 0 | 80.0 | 1495 | 2243 | 1743 | 500 | 0.45 |
| 2 | 79.1 | 1486 | 2229 | 1743 | 486 | 0.44 |
| 4 | 78.2 | 1477 | 2216 | 1743 | 473 | 0.43 |
| 8 | 76.4 | 1459 | 2189 | 1743 | 446 | 0.41 |
| 12 | 74.6 | 1441 | 2162 | 1743 | 419 | 0.38 |
| 16 | 72.8 | 1423 | 2135 | 1743 | 392 | 0.36 |
| 20 | 71.0 | 1405 | 2108 | 1743 | 365 | 0.33 |
The deceleration is visible and it is arithmetic. Over 20 weeks the deficit falls from 500 to about 365 kcal/day, a decline of roughly 27%, with no change in intake. Adaptive thermogenesis would add a further reduction, commonly in the range of 50 to 150 kcal/day during active loss, widening the gap between predicted and observed loss. The table shows the mass-predicted component only; the adaptive component is not included in the deficit column because it is not a fixed value and should not be presented as one.
What people get wrong, and why
The common error is to read the slowdown as evidence that the metabolism has been damaged or broken. That reading is natural because the experience is real: intake is unchanged, yet the scale moves less. The arithmetic explanation is not visible in daily life, because nobody recalculates RMR after each kilogram lost. The body does not announce that it now costs less to run.
A second error is to attribute the entire slowdown to adaptive thermogenesis. The arithmetic effect is larger. In the table above, mass alone accounts for a decline of about 135 kcal/day over 20 weeks. Adaptive thermogenesis, at 50 to 150 kcal/day, is a smaller additional term for most people.
A third error is to treat the 10% standard error on Mifflin-St Jeor as negligible. It is not. A calculated TDEE of 2,243 kcal/day carries a band of roughly 2,020 to 2,470 kcal/day. A deficit of 500 kcal/day sits inside that band, which is why a single calculated figure cannot be treated as exact and why measured expenditure, where available, is preferable.
What follows from this
A deficit is a moving quantity. It can be held roughly constant by reducing intake as mass falls, by increasing activity, or by accepting a slower rate. None of these is a correction for a damaged metabolism; they are responses to a smaller body and to a sustained energy deficit. The rate of loss is a statement about a population, not a promise about an individual, and the individual carries a band around their own maintenance.
This page is general information about energy arithmetic. It is not medical advice and does not diagnose or treat any condition. Anyone with a history of disordered eating, or who is restricting intake in a way that feels out of control, should speak with a clinician or contact a service such as the National Eating Disorders Association helpline (US) or Beat (UK).
Common questions
Does metabolism slow when dieting?
Yes, in two ways. A lighter body costs less to run, which is the larger effect and is predicted by equations such as Mifflin-St Jeor. Adaptive thermogenesis adds a further reduction beyond what mass predicts, commonly in the range of 50 to 150 kcal per day during active loss.
Why am I not losing weight anymore?
The most common reason is that the deficit has shrunk. Maintenance falls as body mass falls, so an unchanged intake produces a smaller gap. Recalculate RMR at the current weight and compare it with intake before assuming anything else.
Is metabolic adaptation permanent?
The evidence does not support permanent damage. Adaptive thermogenesis attenuates with restored energy availability and tends to reverse, partly or fully. The arithmetic effect reverses as mass is regained, because it is a function of mass.
How large is adaptive thermogenesis?
Estimates from controlled feeding studies place it at roughly 50 to 150 kcal per day below mass-predicted expenditure, with wider individual variation. It is smaller than the arithmetic effect for most people and is not a fixed penalty.
How accurate is the Mifflin-St Jeor equation?
It carries a standard error of roughly 10% in adults. A calculated total daily energy expenditure of 2,243 kcal per day therefore has a band of about 2,020 to 2,470 kcal per day. Measured expenditure is preferable where available.
Read next
- BMR: What the Equations Actually Estimate Basal metabolic rate is the energy cost of staying alive. Mifflin-St Jeor, Harris-Benedict and Katch-McArdle give different numbers; here is the spread.
- How Wrong Is Your Calorie Count? Error Sources and Ranges Calorie counting accuracy depends on label error, database variance, portion estimation, and cooking. This page stacks those errors and shows the real range.
- How Big a Calorie Deficit? A Reference Guide How large a calorie deficit should be: derive maintenance from Mifflin-St Jeor, apply 15–20%, and see worked examples at 10%, 20%, and 30% below maintenance.
- Fibre Intake: The Macronutrient Beside Protein Fibre intake targets are 25–30 g daily; typical intake is half that. See how fibre and protein work together, with worked calculations and food sources.
- How Much Protein Per Day: RDA vs. Deficit Targets The RDA for protein is 0.8 g/kg, a minimum to prevent deficiency. For muscle retention in a deficit, evidence clusters at 1.6–2.2 g/kg. Worked examples inside.