Ok to vote down newbies?
Yes: when they don't show any research effort, are unclear, or not useful. Evidencing confusion about concepts† or making incorrect assertions shouldn't in themselves be considered reasons for down-voting a question.
It's wise to presume that new users are
unfamilar with site norms,
inexperienced in asking statistical questions,
& unaware that they can edit their question;
& therefore to explain in a comment how they can improve their question, if someone else hasn't already done so (that goes not only, of course, but especially, for new users). Those with the close-vote privilege will often prefer to vote to close a question when appropriate rather than down-voting it—an edit will automatically place a closed question in the queue for re-opening, whereas an edit to a down-voted question may well be overlooked (see https://stats.meta.stackexchange.com/a/4574/17230).
Do you think this discourages new posters if they have their questions voted down? (Even if perhaps they received answers that are voted up?)
I'd hope they're only discouraged from asking further poor questions. Constructive, polite, criticism should go a long way to mitigate any disappointment at a first attempt at a question's not being well received.
[...] is there some approximate metric, duration of account in days, or minimum rep before you believe a question should be voted down?
No: most poor questions are asked by new users. Down-voting can flag the post for the attention of other users who may help (e.g. users with over 10k rep. can see a list of recent posts with a negative score at https://stats.stackexchange.com/tools), & I'd like to think it discourages premature attempts—perhaps also by new users—at answers. And it can make the difference between a question's being automatically deleted after thirty days, or hanging around for a year, or even indefinitely (see How does deleting work? What can cause a post to be deleted, and what does that actually mean? What are the criteria for deletion?).
† If this muddles the question to the point of making it unanswerable, it's sometimes good advice for the OP to take a step back & explain the problem at hand without recourse to statistical concepts.