Why your weight jumps around day to day
The scale weighs everything on it, and most of what changes overnight isn't fat. Water is the biggest mover: a salty dinner can hold an extra pound or two of it, and the glycogen your muscles store after a high-carb day binds several times its own weight in water. Add food still in transit — meals take a day or two to fully pass through — and it's routine to see swings of one to three pounds between mornings, sometimes more. None of that is fat, and none of it says anything about whether your plan is working. Fat changes on the order of an ounce or two per day even in a solid deficit, which means the daily bounce is roughly ten times louder than the signal you actually care about. Judging yourself by a single morning is reading static.
What a trend line actually tells you
A trend line averages the noise against itself. Water retention pushes some readings up and rebounds push others down, and over enough readings those errors largely cancel — what's left is the slow, persistent movement underneath. That's the number worth reacting to. The trend answers the question the raw scale can't: not "what do I weigh this morning?" but "which way am I going, and how fast?" The rate per week is the practical output. It's the figure you compare against your intentions — losing, gaining, or holding — and it's the input everything downstream uses: the goal timeline tool turns a rate into an arrival date, and because a sustained rate reflects your true energy balance, your trend can also measure your real TDEE — the TDEE calculator's adaptive mode does exactly that from the same kind of weigh-in list.
SMA vs EMA: which smoothing should you use?
5-day average Weighted (EMA)
The 5-day simple moving average (SMA) takes the mean of your last five readings. Every reading gets an equal vote, so the line is steady, easy to explain, and hard to spook — but it reacts to a genuine change of direction only after a few days, because old readings hold their vote until they age out. The exponential moving average (EMA, α 0.1) weights recent readings slightly more while keeping a long memory of everything before them. It turns a little sooner when your direction really shifts, at the cost of wiggling a little more along the way. Honest answer: over a few weeks they tell the same story, and the best choice is the one you'll believe on a bad morning. If a single spike tempts you to panic, take the SMA's stubbornness. If you're changing phases — ending a cut, starting a build — the EMA notices first.
How many days until you can trust your trend
It depends on how noisy your scale life is, and the tool measures that instead of guessing. It fits the trend, looks at the residuals — how far your actual readings scatter around the line — and estimates your personal day-to-day noise. The noisier your readings, the longer a real trend needs to climb out of the static: someone who swings half a pound daily can trust a direction in under two weeks; someone who swings two pounds needs longer. As a rule of thumb, expect ten to twenty days of consistent weigh-ins before the direction deserves your confidence, and treat anything shorter as a rumor. If the wait frustrates you, the fix is consistency, not a better scale — why daily weight changes when the plan is working covers the interpretive side, and consistency beats accuracy explains why the same flawed scale under the same conditions beats a better scale used carelessly.
Reading the scale across a menstrual cycle
Weight moves across a cycle, and the movement is fluid rather than tissue. What is less widely known is how badly the popular version of the story matches the best prospective data. In a year-long cohort of 62 women covering 765 cycles, self-reported fluid retention peaked on the first day of flow — not in the late luteal stretch the premenstrual framing predicts — was lowest mid-follicular, and was associated with neither estradiol nor progesterone. So a curve drawn from the population would be wrong for a great many individual people, which is why this tool does not draw one. Open “I track a menstrual cycle” beside the calculator and it regroups the readings you already pasted, takes your underlying trend out of them, and reports where your heaviest and lightest phases actually fall.
| Peak retention | Lowest retention | Tracks hormones? | |
|---|---|---|---|
| The premenstrual story | Late luteal, in the days before flow | Not specified | Assumed |
| 765 measured cycles | The first day of flow | Mid‑follicular | Neither estradiol nor progesterone |
It also declines to do something you may have seen elsewhere: it does not adjust any calorie figure. The meta-analysis on this question found a small luteal increase in resting metabolism across studies going back to 1930, but among studies published since 2000 the effect was no longer statistically significant, and it sits inside the ordinary 3–5% day-to-day error of measuring resting metabolism at all. A number that cannot be distinguished from measurement noise is not a number worth eating by, so we leave it out and say why.
Sources
- White, Hitchcock, Vigna & Prior (2011), Fluid retention over the menstrual cycle: 1-year data from the prospective ovulation cohort, Obstetrics and Gynecology International
- Benton, Hutchins & Dawes (2020), Effect of menstrual cycle on resting metabolism: a systematic review and meta-analysis, PLoS One
This tool is a reading aid, not medical advice. Rapid or unexplained weight change can have medical causes; if that's what your data shows, a doctor — not a trend line — is the right next step.