Why weight loss stalls, and what to change first
Define it properly first: three consecutive weekly averages in the same band, not a run of bad mornings. When a genuine stall is examined, the usual cause is a deficit that has quietly stopped being kept — intermittent non-adherence reproduces the six-month plateau that clinics see, while metabolic slowdown does not[1]. Adaptation is real but modest, around 92 kcal a day after a large loss[2]. Measure for two weeks before you cut anything.
On this page
Three weeks is the threshold, not three days
Body weight moves for reasons that have nothing to do with stored fat: the water bound up with muscle glycogen, the salt in yesterday's dinner, the fibre and fluid still in transit, and for women the stage of the menstrual cycle. Each gram of glycogen is stored with roughly 3 grams of water[4], so a single high-carbohydrate day can hide a real loss completely.
That is why a stall has to be defined on averages, not on mornings. Take the mean of a week's weigh-ins and compare it with the mean of the week before. If three consecutive weekly averages sit within the same narrow band while you are still eating to a deficit, you have a plateau worth investigating. Anything shorter than that is noise, and treating noise as a plateau is how people cut their food twice for a problem that never existed.
The clock says adherence, not metabolism
There is a timing clue that settles most of this argument before any biology comes into it. Supervised trials and validated energy-balance models both predict that a person eating a fixed, correct deficit keeps losing for one to two years before the arithmetic flattens out. Real clinics see weight stabilise at about six months[1]. The gap between those two numbers is the thing that needs explaining.
When researchers tried to explain that gap, they found that exaggerating metabolic slowdown to implausible levels moved the final weight but never moved the date of the plateau. What reproduced the six-month flattening was intermittent non-adherence: in one long calorie-restriction trial, women were about 80% compliant in the first month and near 40% from the third month onward[1]. A deficit kept four days a week and lost the other three is not a smaller deficit. It is, on average, no deficit.
This is not an accusation of dishonesty. The classic measurement of it found that people who sincerely reported eating little were eating substantially more than they said, and moving substantially less, with intake under-reported by about 47% and activity over-reported by about 51%[3]. Portions drift upward, the oil in the pan stops being counted, weekend meals stop being logged at all. Nobody notices, because each individual omission is small.
Metabolic adaptation is real, and smaller than its reputation
Energy expenditure does fall as you get lighter, and it falls slightly more than body size alone predicts. In one controlled study, participants who lost 14 kg on a very low energy intake over eight weeks showed an adaptation of about 92 kcal per day beyond what their new body size accounted for, and every extra 50 kcal per day of adaptation was associated with about 0.5 kg less weight lost[2].
Take that seriously and keep it in proportion. Roughly a hundred calories a day is a real headwind, worth a deliberate adjustment. It is not a metabolism that has broken, and it is nowhere near large enough to hold a genuine deficit at a standstill for a month.
The other half: your target was built for a heavier person
A calorie target is calculated from the body you had when you set it. Lose weight and the same target becomes a smaller deficit, because a lighter body burns less at rest and less in motion. Nothing has gone wrong; the number has simply expired. Recalculating it after every few kilograms is ordinary maintenance, not a concession. The guide to how your daily calories are calculated shows exactly which inputs change when your weight does.
What to change, in order
Measure before you cut. For two weeks, weigh and log everything, including cooking oil, drinks and what you eat standing in the kitchen. Most plateaus end here, because the gap between what is eaten and what is recorded is where they live.
If the log is honest and the average is still flat, recalculate the target for your current weight. Then add movement rather than subtracting food, since walking and daily activity can be increased without pushing intake toward the floor. Only after that should you reduce calories, and only modestly — the guide to how fast you can safely lose weight explains why there is a floor under the target and what happens when you go beneath it.
And keep the measurements that a stalled scale cannot invalidate: waist circumference, how clothes sit, the weights you are lifting. A month of no movement on the scale while the tape and the gym log both improve is not a plateau in any sense that matters.
Sources
- Thomas DM, Martin CK, Redman LM, Heymsfield SB, Lettieri S, Levine JA, Bouchard C, Schoeller DA. Effect of dietary adherence on the body weight plateau: a mathematical model incorporating intermittent compliance with energy intake prescription. American Journal of Clinical Nutrition 2014;100(3):787–795. ↩
- Martins C, Roekenes J, Gower BA, Hunter GR. Metabolic adaptation is associated with less weight and fat mass loss in response to low-energy diets. Nutrition & Metabolism 2021;18:60. ↩
- Lichtman SW, Pisarska K, Berman ER, Pestone M, Dowling H, Offenbacher E, Weisel H, Heshka S, Matthews DE, Heymsfield SB. Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. New England Journal of Medicine 1992;327(27):1893–1898. ↩
- Olsson KE, Saltin B. Variation in total body water with muscle glycogen changes in man. Acta Physiologica Scandinavica 1970;80(1):11–18. ↩