Why the same diet stops working

Three mechanisms, not one

"My metabolism has slowed down" is usually correct and almost always imprecise. Three separate things reduce your expenditure during a diet, they operate on different timescales, and only one of them is what researchers mean by adaptive thermogenesis:

1. You are physically smaller. Immediate, unavoidable, and entirely predictable from the arithmetic.
2. You move less without noticing. Builds over weeks. Usually the largest of the three, and the one nobody accounts for.
3. Your tissue becomes slightly more efficient. The genuine adaptive component. Real, measurable, and smaller than the internet suggests.

Separating them matters because they call for different responses. The first is a recalculation. The second is behavioural and partly fixable. The third you mostly wait out.

A smaller body costs less

Mifflin-St Jeor charges about 10 kcal a day per kilogram of bodyweight, so every kilogram you lose takes roughly 10 kcal off your BMR — and that reduction then passes through your activity multiplier, because moving a lighter body around costs less too.

Someone who set a 500 kcal deficit at 100 kg is running about 355 kcal by the time they reach 90 kg — a 29% reduction in the gap driving their weight loss, with no change in behaviour at all. Their weekly loss falls from 0.45 kg to 0.32 kg for exactly the same effort.

This is not adaptation in any interesting sense. It is a term in an equation. It is also the piece that every fixed-constant projection ignores, which is why those projections drift optimistic over months rather than weeks.

The NEAT drop

NEAT — non-exercise activity thermogenesis — is everything you burn moving that is not deliberate exercise: walking, standing, fidgeting, gesturing, shifting in a chair, taking the stairs. In free-living adults it varies by hundreds of calories a day between individuals, and it is the most volatile component of expenditure by a wide margin.

In a sustained deficit it falls. You take fewer steps, stand for less of the day, and move less within the movements you do make. This happens below the level of decision — people reliably report no change in activity while accelerometers show one. That is precisely what makes it dangerous for planning: it is a real reduction in expenditure that never appears in your log because you never chose it.

This calculator applies it as 5% of your BMR deducted from expenditure, once you have accumulated 30 consecutive days in a deficit. For the 90 kg row that is about 92 kcal a day — roughly the same magnitude as losing 9 kg, arriving on top of it.

Adaptive thermogenesis proper

The narrow definition is the part of the fall in expenditure that remains after accounting for lost tissue — the same amount of body, burning less than it used to. Thyroid hormone output drops modestly, leptin falls sharply, sympathetic nervous system activity decreases, and muscle becomes marginally more efficient at the same work.

In the literature this component typically runs 5% to 15% below what body composition alone predicts, and it is larger after rapid loss than after gradual loss. It is genuinely adaptive rather than pathological: a body that reduces its energy cost during a shortage is functioning correctly.

It is also the component most often overstated. Extreme cases from weight-loss-competition follow-up studies get quoted as though they were typical of somebody losing 8 kg over four months. They are not, and the gap between those cases and an ordinary diet is a matter of degree, duration, and how aggressive the deficit was.

What it adds up to

Stacking the first two mechanisms for the 100 kg profile losing 15 kg: about 218 kcal from the weight term, plus about 89 kcal of NEAT penalty on the reduced BMR, for a total near 307 kcal. A 500 kcal starting deficit is now under 200.

That is what a plateau is, expressed in arithmetic. Not a stall, not a metabolic failure — the deficit closed while intake stayed constant. Set against it, the naive projection is still confidently reporting 0.45 kg a week, and the gap between those two claims widens every week the diet continues.

Run the same figures in the calculator and the two curves on the chart show exactly this: the straight line the classic arithmetic draws, the curve that bends away from it, and the shaded week where the second one flattens out.

What it is not: a broken metabolism

"Starvation mode" as usually described — a body that stops losing fat entirely, or gains it in a deficit — does not happen. Every controlled overfeeding and underfeeding study finds energy balance holding; what changes is where the balance point sits, not whether it exists.

Adaptation is also largely reversible. Expenditure recovers as weight is regained or intake returns to maintenance, with the NEAT component recovering fastest — often within days of eating normally again, which is one of the arguments for planned diet breaks.

What does persist after large losses is a maintenance requirement somewhat below what a never-dieted person of the same size needs. This is one of the better-documented findings in the field, and the practical reading is not that losing weight is futile. It is that maintenance is a distinct phase requiring its own intake figure, rather than something that happens automatically once you stop dieting.

What to do about it

• Recalculate as you go. Every 5 kg lost is worth revisiting your intake target. The figure that produced a deficit at your starting weight will not produce the same one later.
• Protect the NEAT component deliberately. Because the drop is unconscious, a step target is the one countermeasure that works — it converts an invisible variable into a visible one.
• Do not out-run it with a bigger deficit. More aggressive loss produces a larger adaptive response and greater lean mass loss, which makes the next phase harder. See the safe rate guide.
• Eat enough protein and keep training. Neither prevents adaptation, but both preserve lean mass, and lean mass is what keeps your BMR from falling faster than the weight loss requires. The macro guide covers the numbers.
• Plan for maintenance before you arrive. Your intake at goal weight is not your intake before the diet, and discovering that afterwards is how most regain starts.

Sources

The research this guide draws on, and the claim each one supports. Figures that come from the calculator itself — the NEAT penalty, the danger thresholds, the break model — are documented with their own sources on the methodology page.

1. The 10 kcal per kilogram weight termMifflin MD, St Jeor ST, et al. (1990). A new predictive equation for resting energy expenditure in healthy individuals. American Journal of Clinical Nutrition 51(2):241–247.
2. NEAT, and how widely it varies between peopleLevine JA (2002). Non-exercise activity thermogenesis (NEAT). Best Practice & Research Clinical Endocrinology & Metabolism 16(4):679–702.
3. Expenditure falling further than the loss of tissue alone predictsLeibel RL, Rosenbaum M, Hirsch J (1995). Changes in energy expenditure resulting from altered body weight. New England Journal of Medicine 332(10):621–628.
4. The size of adaptive thermogenesis and the hormonal changes behind itRosenbaum M, Leibel RL (2010). Adaptive thermogenesis in humans. International Journal of Obesity 34(Suppl 1):S47–S55.
5. The extreme case: expenditure six years after The Biggest LoserFothergill E, Guo J, et al. (2016). Persistent metabolic adaptation 6 years after "The Biggest Loser" competition. Obesity 24(8):1612–1619.
6. A lower maintenance requirement persisting after weight is lostRosenbaum M, Hirsch J, et al. (2008). Long-term persistence of adaptive thermogenesis in subjects who have maintained a reduced body weight. American Journal of Clinical Nutrition 88(4):906–912.