Self-answering is hard to do well
It's hard to write a good question when you're planning on self-answering, and a few of our 5e questions suffer from this. It's even harder to write a good answer to such a question. If the problem as-written boils down to "X in 5e is different from [other edition], help help", whatever other virtues the question might have are going to be dragged down by the lack of specifics presented, and the resulting answers are going to suffer the same.
The general problem with self-answered questions is that often the problem is not well-explained, because it doesn't need to be: the answerer can read your mind, because you're the same person.
Solution: Unless one is confident in their ability to write a rockin' question even when self-answering, it may be useful to pause a moment: consider not self-answering for a day or two. Having to explain the problem clearly to others makes it easier to avoid subconsciously taking shortcuts while writing the question. To gain that effect, strongly consider writing the question to the community, and letting them have a crack at it first.
Do research to avoid asking about the really low-hanging fruit
Our question about statistics of the monster math is a valid question but suffers a cardinal question-asking sin: it doesn't just lack evidence of research, but it can only exist as a question because of lack of research.
Solution: We're all excited about 5e, but it's doubly important to take a few moments to consider what research can be done toward solving the problem before asking. It might mean the question becomes moot, but if the question remains after research, it's going to be a good question. And those are what we want.
Broad questions are still broad
When a new edition is out, lots of questions we have are pretty broad due to the nature of how we, as humans, approach a new subject: in a top-down manner, engaging first with the high-level overview and then moving downwards into the nitty-gritty details. Right now there will be the temptation to transform our current questions of engagement into posted Stack questions. But many of those are by their nature fairly broad.
This is not terrible or wrong. But it does mean that taking some extra care to consider the question while writing it can make a large difference in quality.
Solution: When trying to come to grips with a large idea, like "how does X work" or "what is different about Y", try to be as specific about your problem as possible. Remember than being general to help the most people is not actually helpful. Instead, ask yourself some questions: Why do I need to know this? What does knowing solve for me? Can I ask for help solving that instead?
If a problem-centric question doesn't result from this introspection, it's possible that the question simply is too broad. In those cases, the solution is to either not post it, or to post it knowing that it will likely get closed.
For a contentious example of broad question, we have the "What terms have new meanings" question. I think this can be significantly improved by moving away from the broad edition-focused question as written, to a more problem-focused question that's implied, using those questions and some hypothetical (but I think, not too inaccurate) answers:
- Why do I need to know this? Because lots of my players are going to assume they know what terms mean.
- What does knowing solve for me? I can focus on the terms that contradict their assumptions when teaching the game to my players.
- Can I ask for help solving that instead? Yeah, actually: I could ask a question like "When teaching 5e to existing D&D players, what invalid assumptions can I expect them to make?" or something like that.
And if rewriting it as a problem-centric question simply doesn't work somehow, it's probably just too broad.