To weigh yourself properly, step on the same scale at the same time of day under the same conditions — in the morning, after using the bathroom, before eating or drinking — log the number, and then look at the trend line rather than the reading. The goal of a weigh-in is not to get a "true" number on any given day; it is to feed a series of comparable numbers to the smoothing that finds the direction underneath.
Same time, same conditions
Body weight varies through the day by more than most people expect, and most of that variation is water and food rather than anything you would want to track. A weigh-in protocol simply fixes the variables you can fix.
| Variable | What to hold constant | Why it matters |
|---|---|---|
| Time of day | Morning, on waking | Overnight, intake and activity are the same every day — the closest thing to a controlled condition |
| Bathroom | After urinating, before eating or drinking | Removes the most variable morning mass |
| Clothing | None, or the same light clothing | Clothes are a fixed offset only if they are the same clothes |
| Scale | The same scale, same spot, hard level floor | Different scales read differently; carpet and tilt change the reading |
| Posture | Stand still, weight even, hands free | Leaning or holding a phone shifts the load cells |
MacroDiet's in-app card says it in one line: weigh yourself under similar conditions each time; in the morning, after urinating, works for most people. The exact conditions matter less than repeating them. A trend line copes fine with a fixed offset (you always weigh in a T-shirt) and badly with a varying one (T-shirt Monday, jeans Tuesday, after lunch Wednesday).
In plain terms: make every weigh-in the same kind of measurement, and let the maths handle the rest.
What moves the number day to day
Between two mornings the amount of fat or muscle in your body changes by grams, well below what a scale can resolve. What changes by hundreds of grams or more is everything else:
- Glycogen and its water. Stored carbohydrate in muscle and liver is stored with water. A high-carbohydrate day, or a low one, shows up on the scale the next morning without any change in stored energy.
- Salt and hydration. A salty dinner or a change in how much you drank shifts how much water the body is holding.
- Gut contents. A large or late meal is still partly in transit at 07:00.
- Training. A hard session moves fluid around and changes how much you drank and sweated.
None of this is a problem for tracking, because it is exactly what the trend line is designed to average out (see trend weight vs scale weight). It becomes a problem only if you read the daily number as a verdict.
What to do with an outlier
Log it. The temptation after a heavy weekend is to skip the weigh-in "because it isn't real". Two things argue against that.
First, the smoothing already discounts it. With the 7-day half-life the app uses, a single reading counts for about 9.4% of the trend:
trend = 0.0943 × 82.0 + 0.9057 × 80.0 = 80.19 kg
A reading 2 kg above the trend moves it by 0.19 kg, and if the next mornings come back down the effect fades within a week. This is arithmetic, not a trial result.
Second, choosing which days count introduces a bias the maths does not have. People rarely skip a reading for being suspiciously low. If only the high days are dropped, the trend drifts below reality and the first honest week afterwards looks like a stall. Logging every morning, without judging the number, keeps the line honest — and makes the reading a data point rather than a mood.
If a reading really was an error — you stood on the scale holding a bag — both apps let you fix it: they keep one entry per calendar day, so saving again that day overwrites the first value.
The short version
Weigh in every morning under the same conditions, log the number, and judge progress from the trend line over two to three weeks. Daily swings of a kilogram are water, glycogen and gut contents, not fat or muscle. Log the odd day too: a 2 kg blip moves the 7-day-smoothed trend by about 0.19 kg, and skipping it would bias the line more than including it. Nutrition and training decisions belong on the trend. Educational overview only — not medical advice.