When a red flag was a misunderstanding, and the retraction — that is the question, and I have not found it answered plainly anywhere I have looked.
Posting a small dataset on red flag. It is mine, it is uncontrolled, and the method is stated so it can be discounted appropriately.
What I would like is not agreement but a second dataset collected by someone with no stake in mine. If one exists I would rather read it than argue for this one.
Inability to trace lots: if you ask "where was this synthesised, by whom, under what conditions" and cannot get a clear answer, that is a red flag. Traceability is not glamorous but it is foundational.
That is one dataset and I would not build a rule on it.
Nothing here should be read as an accusation against a named company, and the guidelines exist so that reports remain reports.
Worth reading the earlier posts in this thread before acting on mine.
Adding the measurement that post #4 says would settle it.
Where the red flag discussion usually stalls is that nobody wants to say "I do not know" and everyone is willing to say "it varies". Those are the same sentence with different clothes on.
Post #3 describes the usual case. This is about the unusual one.
If you are new and reading this thread for the answer to red flag: the answer is conditional, the conditions are in the third reply, and the rest of the thread is worth skipping.
Prices well below the rest of the market are a fact rather than a finding. They are worth noting and they do not establish anything on their own.
Inconsistency across documents: a certificate stating one purity, a website claiming another, and a supplier's verbal representation claiming a third, all on the same product, is a red flag for either incompetence or deception.
If anyone has run this properly I would rather read that than my own guess.
The reason red flag is hard to answer is that the obvious measurement and the relevant quantity are not the same thing, and substituting one for the other is silent.
Post #9 and I disagree about the size of the effect, not about the direction.
The guidelines draw that line for a practical reason as much as a fair one: reports accumulate into something useful and allegations do not.
Everything in post #13 holds. The case it does not cover is the one I have.
What I would check first on red flag is whether the thing being measured moved or whether the way of measuring it moved. Those look identical in a graph.
A stock listing that never changes and a lead time that never varies are both worth noticing. Real operations have variation in them.
I had written a reply contradicting post #17 and deleted it. Here is what survived.
The most useful contribution to this subcategory is a resolved case written up with the resolution, because the unresolved ones are all anyone remembers.
Confirming post #17 from a second method, which matters more than confirming it from a second person.
The number people quote for red flag is a central estimate presented without its interval, and the interval is wide enough that the estimate is nearly uninformative on its own.
Narrowing post #18, because the general version has more than one answer.
Where somebody posts a concern, ask what would resolve it before arguing about it. Half of these threads end with a document nobody had asked for.
I would treat that as a working assumption and revisit it.
Testimonials without a lot, a date, a method or a service are not evidence and are not a red flag either. They are simply not evidence.
What I would want before treating red flag as settled: the method, the sample, and whether anyone tried to find the opposite result. Two of the three are usually missing.
That is a cleaner way of putting what I was circling around.
Staff rationale: Hidden by community flags. The substance was fine; the framing attributed motive rather than error, which is R1.
Adding the boring version of red flag, because the interesting version keeps getting posted and the boring one is usually right.
Check the ordinary explanations, in order, and stop when one of them accounts for what you are seeing. Most of the time the second one does.
The corresponding entry is in the public moderation log. Hidden posts are never deleted.
Post #27 describes the usual case. This is about the unusual one.
Inconsistency across documents: a certificate stating one purity, a website claiming another, and a supplier's verbal representation claiming a third, all on the same product, is a red flag for either incompetence or deception.
That is the shape of it. The detail is where I would expect to be corrected.