Why does your weight fluctuate so much, and how do you read the scale?

What a kilo overnight is made of

Your body is more than half water. In an adult man it’s about 58 percent, which for an 80 kg (176 lb) person is roughly 46 liters, and a liter of water weighs a kilo. So if your body holds on to two percent more of it on Tuesday than it did on Monday, that’s close to a kilo on the scale with not a gram of fat involved.

And water is nearly all of what moves. A man weighed himself every morning for almost thirty years, 9,521 mornings in all, undressed, after the bathroom, before breakfast. His weight drifted from 75 kg to 84 kg over those decades, but from one morning to the next it bounced by about half a percent either way, and it did that in the years he was gaining, the years he was losing and the years nothing changed. The physiologists who analyzed the record put it plainly: a change of that size in fat, protein or anything else solid inside a single day isn’t realistically possible. It has to be water.

A bigger group says the same over a longer stretch. When 46 adults who weren’t trying to change anything were weighed two weeks apart, 84 percent of whatever their weight had done in between was fat-free mass, which mostly means water. Only about a sixth of a two-week swing was fat.

So where does the water come from, and where does it go? Three places, and you can feel all of them.

The food in you. Everything you eat and drink has a weight, and it stays with you a lot longer than the meal does. The time from a mouthful to the toilet has a median of about 60 hours, roughly 55 in men and 72 in women, measured in 220 British adults. Part of Monday’s dinner can still be on board on Wednesday morning. A big plate of vegetables and a lot of fiber add to what’s in transit, which is one reason a new, healthier diet can push the scale up for a few days before it goes down.

The carbohydrate your muscles store. Your liver and muscles keep a reserve of carbohydrate, called glycogen, and every gram of it is stored with three to four grams of water. Eat a lot of carbohydrate for a couple of days and the reserve fills up; it’s why athletes who carb-load before a race gain about a kilo to a kilo and a half, nearly all of it water. Eat very little carbohydrate, or train hard, and it drains, water included. In one trial, 22 men who ate a moderate amount of carbohydrate for three days weighed 0.7 kg less than when they ate a lot, and performed exactly the same.

The salt in your food. Salt holds water with it for a day or two while your kidneys clear it, and the researchers who study daily weight list it as one cause of the noise. What it doesn’t do is pile up. In one careful study, 32 men on a metabolic ward went from a low-salt diet to about ten times as much, and their blood volume rose while their total body water and their weight stayed where they were. So the salt in a pizza is part of why the next morning is heavier, and probably a smaller part than it gets credit for. The pizza itself, and the carbohydrate in it, are the bigger parts.

That’s the list, more or less: water, moved around by what you ate, and the food itself. None of it is fat, and most of it passes within a few days of normal living.

How much swing is normal

The thirty-year record put the morning-to-morning swing at about half a percent of body weight, and smaller groups measured over a week land in the same place: in 86 young men weighed every morning from Monday to Friday, the typical change from one morning to the next was 0.4 kg, on about 75 kg. Across a week rather than a day, the swing in the long record grew a little, to about 0.7 percent.

So for an 80 kg (176 lb) person, roughly 0.4 kg (about a pound) either way is an ordinary morning. For a 60 kg (132 lb) person it’s 0.3 kg; for a 100 kg (220 lb) person, 0.5 kg. And there’s the whole problem in one line: a sensible rate of fat loss, the one we aim for with most clients, is about half a percent of your weight a week, which on 80 kg is also 0.4 kg. The noise of one morning is the same size as a whole week of progress, so a single weigh-in can’t tell you which one you’re looking at.

On top of the random part sits a rhythm. When 1,421 adults across Denmark, Portugal and the UK weighed themselves on smart scales for about a year and a half, their weight was highest on Monday, fell through the week to a low on Friday, and climbed again over the weekend, by about a third of a percent from low to high. A smaller American study found the reason in the diaries: more food on Saturdays, less moving about on Sundays. So for most people a Monday reading sits on the week’s high point, which is worth knowing before you let it decide anything.

Christmas does the same thing at a bigger scale. The same European group gained about 1.3 percent of their body weight between the first week of December and the first days of January, around a kilo for most of them, and by March a quarter of it was still there. That one isn’t all water, and it isn’t what this page is about: it’s two weeks of eating more, and it comes off the way any gain does, slowly.

The normal swings, sized to you

Those are averages, and you aren’t one. The useful thing is the size. Half a percent is what the ordinary noise looks like, not a line: a smaller move might be noise or the start of something, and one reading won’t tell you which. A bigger move, a kilo overnight, usually has a cause you can name, and it’s probably on this list.

Why you’re heavier after eating out

Put the pieces together on a Friday. You go out, eat a proper dinner, share the bread, have a dessert and two drinks. Saturday morning you’re 1.2 kg (2.6 lb) up, and you can feel it. Most of the meal and the drinks are still on board, because your gut is a two-day pipeline, not a two-hour one. The bread, the pasta, the dessert and the beer refilled your carbohydrate reserve, water included. The salt made you thirsty, so you drank more than you would have at home. And the fat part is only what the extra calories could have made, which even on a big night is a small fraction of a kilo, and the scale can’t tell you which part of the number it was.

Over the next few days the pipeline empties, the reserve settles and the salt is cleared, and the number drifts back toward where it was. The mistake isn’t going out. It’s weighing yourself on Saturday morning and letting that number tell you something about Friday night that it can’t.

The first week of a diet, and the Monday after a weekend off

The reverse happens when you start eating less, and it fools people the other way. Cut your calories and your body draws on its carbohydrate reserve first, water and all, which is why the first week of a diet tends to drop faster than the weeks after it. In the classic study of this, six people with obesity spent ten days on an 800-calorie diet, twice: once with ordinary carbohydrate in it and once with almost none. On the low-carbohydrate version they lost 467 grams a day, and 61 percent of it was water. On the ordinary version they lost 278 grams a day, and 60 percent of that was fat. Same calories, nearly twice the weight loss, and the extra was water leaving with the glycogen.

That’s most of the story of the famous first week, and of the famous second week, when people who dropped two kilos drop nothing and decide the diet has stopped working. More often, week one was the reserve draining, and the weeks after it are the better guide to the real pace, which is slow. The researcher who worked out the glycogen numbers in the 1990s put the two illusions in his title: easy weight loss at the start, and excessive regain the moment you eat carbohydrate again.

The regain half is the Monday after a weekend off. Two days of normal eating refill some of what the diet drained, and the scale hands back a number that was never fat. If you weigh every day, weigh that Monday too; just don’t read it on its own. It goes into the week’s average, and the average does the judging.

In our coaching the first week is where we put the scale down on purpose. We have people change nothing for a week and simply log what they eat, and most lose a little weight that week anyway, because paying attention changes what you do. We also give any plan at least four weeks before we judge it by the scale, for exactly the reason this page exists: the number moves for reasons that aren’t fat.

Your period, and the week after it

If you have periods, there’s a monthly wave under all of the above, and it’s bigger than most people expect and smaller than the stories. In 42 women weighed twice a week for a whole cycle, in the morning after a twelve-hour fast, weight was about half a kilo higher on the first day of the period than a week later, and the difference was water outside the cells. Waist, skinfolds and fat mass didn’t change. The weight then crept back up through the month to peak at the next period. A second, smaller study of active women found the same shape with a smaller gap, about 0.3 kg, and put it down to hydration shifting rather than tissue.

So a jump in the day or two before your period is far more likely to be the cycle than the diet, and in the women measured, the days after the period were the lightest part of the month, which makes them a fair point to compare against. Compare the same week of the cycle with the same week last month, or look at the monthly average and let the rest go. The studies above were on women who weren’t using hormonal contraception; if you are, your own pattern is the one to learn, with the same monthly comparison.

Training, sweat and creatine

Exercise moves the scale in both directions in the same week, which is why people who start training and dieting at the same time get the most confusing readings of all.

Down first: sweat. The way sports scientists measure how much someone sweats is to weigh them before and after a session, because the difference is almost all water. The point at which they call it too much is a loss of two percent of body weight, 1.6 kg on 80 kg, and a long, hard session in the heat can get there. The day’s drinking puts it back by the next morning, if you drink. Weigh yourself straight after a long run or a sauna and you’ll see a number that says a lot about sweat and nothing about fat.

Then up: water in the muscles. Start lifting, or come back after a break, and your muscles swell with fluid in the first weeks. In men returning to leg training after six months off, the thigh measured about three percent bigger at week three, and a good part of that early increase was swelling rather than new muscle. So if you started lifting three weeks ago and you’re a kilo heavier, the likeliest explanation is water in the muscles you’re training, which fits with the training doing its job. A kilo of new muscle doesn’t come in three weeks, and a kilo of fat would need a surplus you’d probably know about. How long the real thing takes is its own question, and our article on how long it takes to build muscle answers it.

Creatine does the same on purpose. In the first week of taking it, the muscles hold roughly half a liter to a liter of extra water, and that’s where the half kilo to kilo on the scale comes from. Weight gain is the one side effect the research on creatine reports consistently, and the early gain is water.

Can a bad night’s sleep change the scale?

Not much on its own, as far as the research goes. What a short night clearly changes is the next day: when researchers pooled eleven trials that cut people’s sleep short and then let them eat as they liked, the short sleepers ate about 385 calories more the next day and burned no more, a pastry’s worth of extra for every night it continues. Do that for a week and it adds up, through what you eat rather than through the night itself. On a diet, short sleep also changes what the weight you lose is made of: in a study where ten people lost the same three kilos on the same diet twice, with long nights more than half of it was fat, and with short nights a quarter. That’s the sleep side of this, and if yours has slid, our article on fixing a sleep schedule is where to go.

How to weigh yourself, if you weigh yourself at all

Pick one set of conditions and keep them. The studies on this page weighed people first thing in the morning, after the bathroom, before food or drink, on the same scale, and that’s the easiest routine to keep, because it’s the moment of the day with the least in you. Any fixed time and routine can work; what doesn’t is 7 PM on Monday against 7 AM on Thursday, because then you’re comparing two different things.

Then stop reading mornings and start reading weeks. Think of the little frames-per-second counter in the corner of a game. It flickers between 55 and 70 every second, and nobody decides a game runs badly because of one dip to 48; you watch it for a minute. The scale is the same kind of number, read once a day, so you watch it for a week: add up the seven mornings, divide by seven, and compare that with last week’s seven. One week against the last can still fool you, especially around a period or a weekend away. Two or three weekly averages in a row pointing the same way are a much better guide.

Every morning or once a week? Our answer hasn’t changed since the calorie deficit article: every morning if it doesn’t mess with your head, once a week if it does. The research behind that is honest in both directions. In adults who’ve signed up to lose weight, trials of daily weighing haven’t found more anxiety, lower mood or more binge eating, and the daily weighers tend to lose a little more; in one six-month trial of 91 overweight adults, body satisfaction went up, including for the people who didn’t lose weight. But in young women who weren’t trying to lose weight, two trials found that daily weighing made moods swing more and self-esteem drop. So it’s a tool, not a duty. If the number sets your mood for the day, weigh once a week, or not at all, and use the tape measure below.

If you weigh once a week, the same idea applies with one change: a single reading carries a whole morning’s noise, so pick the same morning each week, not the one straight after the weekend, keep the conditions the same, and judge the trend across four or five readings rather than against last week alone. Expect it to zigzag. A direction you can see over a month is worth acting on; one reading that’s up is one reading.

One more thing to ignore: the body-fat and muscle-mass numbers on a smart scale. They’re worked out from how easily a tiny current passes through you, which depends on how much water you’re holding, which is the very thing that moves. In the 86 young men weighed every morning for a week, the scale’s body-fat reading wobbled by more than a kilo for the same person between Monday and Friday, while nothing about them had changed. The weight reading is fine. The body-fat reading isn’t telling you anything you can use from one week to the next.

If the scale isn’t moving, is anything happening?

Often, yes. Everything above can hide a week of fat loss, and some of it can hide two: a period, a few days of more carbohydrate, the water that arrives when you start training, a salty week, the morning after a long drive with no stops. Half a percent of fat gone and half a percent of water arrived reads as zero. That’s not the plan failing. That’s the scale weighing two things at once.

So keep a second gauge, and the best one costs nothing: a tape around your waist. It isn’t immune to the noise either. It moves with what’s in your gut, with bloating and the cycle, and with how you stand and breathe, and even a careful measurer gets a slightly different number on a repeat, by a centimeter or so. Its strength is the slow picture: fat loss shows up there over months, including in stretches when the scale is stuck. So once a month, same spot (halfway between your lowest rib and the top of your hip bone, or the top of the hip bone itself, as long as it’s the same one every time), tape level, after you breathe out, no sucking in, and read it the way you read the scale, as a trend across several readings rather than a verdict from one. Clothes do the same job for free: the belt notch, the jeans that button without a fight. And the signs we tell our clients to watch are the ones no scale can read: are you stronger, is there less pain, does the mirror show it, do your clothes fit better, and how are you starting to feel.

If you’re dieting and want the full method for telling it’s working, the lifts, the tape and the scale read together, the calorie deficit article has it in its last section.

And when none of the gauges move for a month, three or four weekly averages flat and the tape flat with them, that’s longer than the ordinary swings tend to run, and a reason to look at the plan again rather than wait. What to look at is a different question with different answers. Until then the rule we use is the one we give every client: give a change four weeks before you judge it, and judge it on all the gauges, not the scale alone. If you’d rather have a coach reading the numbers with you, that is what we do at Gains For Gamers.

When the scale is a doctor’s question

Two patterns don’t belong on this page. Weight that climbs over a few days and stays there, with swollen ankles, legs or belly, is a reason to see a doctor, because it can be fluid your body is holding for a medical reason. If you’re also short of breath doing everyday things or when you lie down, don’t wait for an appointment: call your doctor or the urgent care line the same day, and if you’re struggling to breathe, call emergency services. And weight that keeps falling when you aren’t trying to lose it is a doctor’s question too, not a diet one. The rest of this page is about the ordinary swings. These two aren’t ordinary.