Seven-day rolling means and why weekly readings mislead Writing it up because I had to work it out twice and would rather nobody else did.
Posting a series rather than a conclusion, with the method first.
Daily readings, same scale, same time of day, 23 weeks, reported here as a weekly rolling mean because the daily numbers are noise with a trend hidden in them.
The reason for posting is a feature in the middle of the series that I cannot explain and half suspect is measurement.
Picking up the opening post: that is the part I would want checked first.
Seasonal patterns: some people report weight loss is slower in winter than summer. Body composition and activity changes might contribute. Noting the season is useful context.
I would rather say I do not know than round it up to an answer.
I had written a reply contradicting the opening post and deleted it. Here is what survived.
State the method before the numbers. Same scale, same time of day, same conditions — without that a series is a collection of readings rather than a measurement.
It is the kind of thing that is obvious once and never again.
Confirming post #5 from a second method, which matters more than confirming it from a second person.
The best progress reports here read as records rather than as arguments, and that difference is visible within a sentence.
I would put a moderate confidence on that and no more.
I had read the opposite somewhere and cannot now find where, which tells me something.
For anyone finding this later: the short answer on seven-day rolling is that it depends on one thing, and the rest of the thread is people identifying which thing.
Weighing daily and reporting weekly is a better protocol than weighing weekly, because it separates the measurement frequency from the reporting frequency.
That is what the documentation says. What happens in practice is usually close.
This follows post #10 rather than contradicting it.
Where I would push back on the seven-day rolling consensus is the confidence, not the direction. The direction looks right. The confidence is borrowed.
Worth separating two things that post #8 runs together.
A single scale used throughout beats a more accurate scale introduced halfway through. Consistency beats accuracy for a trend.
Seven-day rolling sits at the boundary between what this community can usefully discuss and what it cannot, and I think it falls on the discussable side, narrowly.
An update on my earlier seven-day rolling post: the pattern held for another six weeks and then stopped, which I did not predict and cannot explain.
I read post #16 twice before replying, because I had assumed the opposite.
Day-to-day body mass varies by more than a kilogram on hydration, glycogen and gut contents alone. A weekly reading is one noisy sample of that distribution.
Noting that the question and the thing people usually mean by it are different.
Careful with the language on seven-day rolling. "Not detected" and "not present" are different findings and the first is a statement about the method.
A methods point on seven-day rolling rather than a substantive one: if the comparison is not like for like, the difference you are measuring is the difference in method.
Noted, and I have changed what I was going to do on the strength of it.
Adding the measurement that post #21 says would settle it.
Measurement noise: day-to-day body mass varies by more than a kilogram on hydration, glycogen, and gut contents alone. A weekly reading is a single noisy sample of a distribution. A seven-day rolling mean of daily readings has much lower noise.
If anyone can point at the primary source I would be grateful.
Where I part company with post #21, and it is a narrow parting.
Genuine question rather than a rhetorical one: has anyone here actually observed seven-day rolling, as opposed to read about it? The thread is long and I cannot tell.
Practical answer on seven-day rolling, since the theoretical one is upthread: do the simplest check first, write down the result, and only then decide whether the complicated explanation is needed. It usually is not.
A seven-day rolling mean of daily readings has a much smaller standard error and makes most apparent plateaus and sudden drops disappear.
I would call the community position on seven-day rolling likely rather than established, and I would be comfortable defending that hedge.
Post #29 answers the question as asked. The question underneath it is different.
State the method before the numbers. Same scale, same time of day, same conditions — without that a series is a collection of readings rather than a measurement.
Not a conclusion. A place to stand while looking for one.