Accepted answer
Over 28 days, expect one and a half to two and a half kilograms of it to be water, and on a fast start that can be most of the total. The arithmetic is glycogen. A well-fed adult stores roughly 400 to 500 grams of it, each gram bound to about three grams of water, so a substantial depletion releases 1.2 to 1.5 kg of water from that alone. Add a fall in sodium intake — which is what happens when total food volume drops — and the extracellular compartment contracts by a few hundred grams to a kilogram more. That is why week 1 flatters and week 4 disappoints: the fluid is a one-off transfer, and once it is gone the curve reverts to whatever the energy deficit alone supports, which is roughly 0.5 to 1 kg a week. It also runs in reverse. A single high-carbohydrate day refills glycogen and its water, and the scale reports two kilograms of "regain" that contains no fat whatsoever. The measure that is not fooled by any of this is a waist circumference taken the same way each week. It moves slowly, and slowly is the point.
Before changing anything, confirm the plateau is real using a trend rather than a reading.
Diet breaks — a fortnight at maintenance — have modest evidence for improving adherence and none for accelerating loss, which is still a reason to consider them.
What each body-composition method measures
| Method | Measures | Sensitive to | Least significant change |
|---|
| DEXA | Three-compartment by attenuation | Hydration, positioning | ~2–3 % regional lean |
| BIA (consumer) | Impedance, modelled | Hydration, food, temperature | Not usable at this timescale |
| Air displacement | Two-compartment by density | Lung volume, hair, clothing | ~1–2 % fat mass |
| Tape and scale | Circumference, mass | Technique | Surprisingly usable as a trend |
Intake drift is real and invisible: portion sizes creep, untracked items accumulate, and self-reported intake is systematically underestimated even by people who are trying hard.
Adaptive thermogenesis beyond that predicted by mass loss is documented in metabolic-ward and doubly labelled water studies and persists for years in some cohorts.
A plateau on a scale is not a plateau in body composition; a scan may show change the scale does not.
Measure intake honestly for a week before changing anything.
5The point about protein being hardest to eat exactly when it matters most is well made. – triple_agonist_q 5 months ago add a comment