Why the scale lies, and what to read instead

You cannot gain a kilo of fat overnight. Here is what you actually gained, and the number to read instead of the one on the display.

Around 6 minutes · Weight

You did everything right on Monday. You weighed on Tuesday and you were a kilo heavier. Nothing you ate on Monday could have done that, and yet there it is on the display, and it is very hard not to read it as a verdict.

It isn't one. Here is the arithmetic that settles it.

That figure is a rule of thumb rather than a constant — real tissue is not pure fat, and the number moves depending on who is doing the measuring — but it is the right order of magnitude, and the order of magnitude is the whole point. Whatever the scale showed you this morning, almost none of the overnight change was fat.

What the extra kilo actually was

Bodyweight is mostly water, and water moves around for reasons that have nothing to do with fat.

• Glycogen. Your muscles and liver store carbohydrate as glycogen, and each gram of it is held with roughly three grams of water. Eat a carb-heavy meal after a low-carb stretch and you refill those stores — which is a real gain in mass, and none of it is fat. Go the other way and you "lose" a couple of kilos in three days, which is why low-carb diets feel miraculous in week one.
• Salt. A saltier day than usual means more water retained until your kidneys catch up. A takeaway can do this on its own.
• Food still in transit. What you ate yesterday has mass and has not finished its journey. A high-fibre day leaves more of it in you for longer.
• New or harder training. Muscle you have just worked holds extra water while it repairs. Starting to exercise can put the scale up for a fortnight while you are actively losing fat.
• Hormones. For anyone who menstruates, fluid shifts across the cycle can comfortably exceed a kilo and follow a monthly pattern that has nothing to do with the diet.
• Sleep and stress. Both move fluid balance. A bad night can cost you a morning's apparent progress.

Any two of those can stack to a kilo or more. All of them can reverse within days. None of them is the thing you are trying to measure.

The signal is smaller than the noise

This is the part people find genuinely surprising. A sensible rate of fat loss is somewhere around 0.5% of bodyweight a week — about 400g for someone at 80kg. That is what a good week looks like.

Daily water swings are routinely two or three times that. So on any given morning the number in front of you is mostly noise, with a week of real progress hidden somewhere inside it. Reading a single weigh-in is reading the noise.

Weigh more often, not less

The intuitive response is to weigh less — once a week, to avoid the noise. It is the wrong move, and it is wrong for a reason worth understanding.

Weighing weekly does not remove the noise. It gives you one sample that contains all of it. If your Wednesday happened to follow a takeaway, that is your whole week's data point, and next Wednesday will be compared against a number that was never true.

Weighing daily and averaging does remove it. Seven readings around a true value average out much closer to that value than any one of them sits. The noise is random; the fat loss is not. Given enough samples the random part cancels and the trend is what's left.

That average is what people mean by trend weight. It is the line through the scatter, and it is the only weight figure worth making decisions on.

What that looks like in practice

• Weigh every morning, after the loo, before anything else, no clothes on.
• Record it and then stop thinking about it. The daily number is a data point, not a result.
• Read the trend line, not today's reading.
• Judge progress over three to four weeks, not three to four days.

The consistency matters more than the time of day. Same conditions every morning means the errors are at least the same shape, and a consistent error cancels out of a trend where a random one does not.

Reading a trend that has stalled

Once you are reading a trend rather than a day, a flat fortnight actually means something — and it still might not mean what you think. Four weeks is the honest window for deciding the trend has changed, because water can mask real fat loss for a surprisingly long time. The classic version is starting a new training programme: the fat is going, the water arriving to repair the muscle roughly cancels it out on the display, and then a few weeks later the whole lot shows up at once. Nothing happened that week. It had been happening all along.