I like to contribute to MO by flagging questions that I think are rather general and thus more suitable for e.g. Mathematics.SE. Recently it is becoming clear to me that my idea of "research level" is not always the same as that of the moderators. To make my flagging more accurate (yielding a higher "accepted" rate for me and less needless work for me and the moderators) I would like to get a better understanding of when I should (not) flag in practice.
For a concrete example let us consider Does this geometry theorem have a name?. I do not know the answer, yet it seems like a basic question about Euclid's geometry, which to me is undergraduate or even high-school level mathematics. Therefore I would guess the OP has a higher chance to get an answer at Mathematics.SE (which I expect to have many more active users) or perhaps "History of Science and Mathematics".SE or "Mathematics Educators".SE (whose users might be more familiar with classical geometry à la Euclid).
However, both (quite a few!) voters and the moderator(s) that handled my flag disagree with me. In fact, these two seem to be correlated: I am under the impression that flags for questions with a positive rating by votes tend to take longer to get accepted/declined, and are usually declined.
Should I avoid flagging questions that I think are not research level when they have a positive rating by votes? Should I perhaps downvote such a question instead?
I understand that the latter mean that at least some users think it is a relevant question, so it might be acceptable even if it would not be research level. (So far I have avoided downvoting in such cases since I do not think the questions are bad per se, but I just think they should be posted elsewhere instead.)