Why does my weight change 5 lbs in a day?

You weighed 178 on Friday. Monday morning it’s 182. You ate normally. Nothing’s wrong with you and nothing’s wrong with the scale; you just watched water move. Here’s what actually happened, with the numbers.

Most of what moves day to day isn’t fat

In 2017 a team at Wisconsin tracked 46 adults for two weeks with no diet, no exercise plan and no instructions, then measured what their weight change was made of. Their weight moved by 0.26 ± 1.2 kg over the fortnight, and 84% of every unit of change was fat-free mass, which over two weeks means water. Sixteen percent was fat. A short-term swing of one to three kilos carried about 2,380 kcal per kilo, a third of what pure fat would.

So when you see a four-pound jump, somewhere around three and a quarter pounds of it is water and lean tissue shuffling around. The fat you actually gained or lost is buried under that, and you can’t see it from one reading.

Bhutani S, Kahn E, Tasali E, Schoeller DA. Physiol Rep 2017;5(13):e13336. doi:10.14814/phy2.13336

Where the water comes from

Three things do most of it.

• Glycogen. Your muscles and liver store carbohydrate as glycogen, and every gram of it is stored with at least 3 grams of water. A low-carb day empties some of that and the water goes with it. A pizza night puts it back. That’s pounds moving with no fat involved either way (Murray & Rosenbloom, Nutr Rev 2018).
• Salt. In the DASH-Sodium trial, 412 adults ate identical calories at three sodium levels for four weeks each. On the saltiest level they weighed a little more, on the same food (Juraschek et al., Hypertension 2020). And your body actually conserves water when salt goes up, rather than flushing it, which is why a salty dinner shows up the next morning (Rakova et al., J Clin Invest 2017).
• Timing. A late meal, a big drink before bed, or a hard workout that leaves you dehydrated all land on the next reading. None of them are fat.

Your weight has a weekly rhythm, and it isn’t your imagination

Eighty adults logged 4,657 weigh-ins on connected scales for anywhere from 15 to 330 days. Their weight peaked at the start of the week and bottomed out at the end of it, reliably enough to publish.

Orsama AL, Mattila E, Ermes M, van Gils M, Wansink B, Korhonen I. Weight Rhythms: Weight Increases during Weekends and Decreases during Weekdays. Obes Facts 2014;7(1):36–47. doi:10.1159/000356147

Which means your Monday number is the worst possible day to judge yourself on, and it’s the day most people do it. It also means a week of readings catches one weekend and can’t tell you whether the weekend was the pattern or the exception. Two weeks can.

What about the menstrual cycle?

You’ll read “2 to 5 lb of water retention” on every women’s health site. I used to repeat it too. The one recent study that measured body composition at four points across the cycle, in 54 people, found no significant cycle effect at all (Stein-Brüggemann et al., Eur J Sport Sci 2025). Some people clearly do see a swing around their cycle. How big and how consistent it is actually isn’t settled, which is why the method gives you a marker for it rather than an assumption about it.

So what do you do with a number that won’t hold still?

You stop reading single frames. Three things, in order:

1. Take two readings, not one. Right after your last meal, and on waking before breakfast or water. The evening number is your ceiling, the morning number is your floor, and the gap between them is the only part that’s about your body rather than about your dinner.
2. Average the gap over 14 days. That covers two weekends and roughly half a cycle. One night is noise. Fourteen is a pattern, and the average is a number you can actually plan with.
3. Watch the trend of the ceiling, not the floor. The overnight drop is mostly water. The fat shows up slowly, in how your evening number drifts down over weeks. Plot a 7-day average and the noise separates from the signal on its own.

One more reason not to trust a single high or low reading: an unusually extreme number tends to be followed by a less extreme one for purely statistical reasons. Epidemiologists call it regression to the mean, and it fools people every week (Barnett et al., Int J Epidemiol 2005). A plan built on your best-ever morning is a plan built on a fluke.

Want to see your own gap? Type in last night’s and this morning’s weight. It takes ten seconds and stores nothing. Then ask yourself: over the last month, how many of the numbers that upset you were Monday numbers?