I was the first close voter for this question: Can DFT be considered an ab initio method?. But I have cast a re-open vote now. The question has two good answers (one of them is deleted, because I told the user to delete it).
The reason I voted for closure was because I didn't quite know what I was doing back then. It was thirteen days into Private Beta and I knew a lot less then than I do now (and even now I know a lot less than I will know in the future). Specifically, I didn't know that cross-site postings are allowed until Rashid commented on that specific question. Instead, I thought it was extremely frowned upon to post something that has already been posted elsewhere, because it might show a lack of research before asking the question, and I was even more than worried than usual about that at the time, because we were in the middle of a Private Beta evaluation period where we'd be shut down if we didn't demonstrate the ability to run the site properly.
Looking back now, I welcome everyone's opinion on whether or not we should allow cross-postings (the consensus on Meta.SE is that it's allowed, but some specific sites like Chem.SE do not allow it, which is perfectly up to the relevant community), but in general I want to address the question of whether or not we really need questions to be closed when there's already a good answer.
What value is there in having a question closed?
In my opinion it is that it helps to avoid people possibly wasting their time answering something that was asked somewhere else already (in the case of duplicate questions), or to reduce the size of our unanswered queue so that it's more "manageable" (in the case of abandoned questions and off-topic questions) or in very rare cases for this site, to uphold the quality standards of posts on the site (in the case of spam or very low-quality questions, i.e. questions that you have already flagged to be VLQ).