I'd like to push back against the thought that closure is a bad thing.
We're far from the only SE site to have this problem. It's part of what comes with the territory as the site grows. We'll get new users that don't bother reading, think this is a forum like everybody else, and ask their cool questions. Most often, this will result in a closure, as their cool question either hits one of our blanket bans (which, as @badp stated, is something we've tried and found we're crap at doing), or have good reason to close them.
This new user is used to forums, where closure is permanent and unchallengable. Here at Arqade, though, closure isn't a permanent state (unless it's a duplicate). Even then, it can be re-opened if it's sufficiently different.
If anything, closure is a good thing, not something to do in the case of last resort. For clearly off-topic questions, it keeps the signal to noise ratio nice and high. These are the blanket bans.
For the rest, one of two things will happen:
- The question gets closed, and eventually deleted.
- The question gets fixed, and re-opened.
Either one strengthens the signal ratio, and that's a very good thing.
For new users, we strive to leave comments as to why their question got closed, so it can be fixed (if it CAN be fixed). Sometimes we get meta posts about them. This brings more attention to the questions than if we had just left them alone, and usually identifies and fixes any problems the question might have. This is the whole reason we HAVE those other close reasons. It identifies the part of the question that isn't acceptable so that the asker (or others) can fix it.
So closing isn't a bad thing; it's gained a bad rep from forums. New users won't like it, but there's a reason we have this ability; to determine what does, and doesn't belong, and fix that which might be salvagable.
The other part I'd like to address is the new user experience. Are we welcoming to each and every new user? Probably not. It's something we're working on, and the comments on questions is one aspect of being nicer.
But I'd like to challenge the notion that it is entirely our responsibility to keep those new users. We're a community, with set rules in place. A barrier to entry is to adapt to the community's rules and quirks. It's not up to us to coddle and baby new users in order to get them to stay. It's as much their responsibility to conform to our standards as it is for us to be open to welcome new members. The new user has to want to be here, rules and all.
Trying to keep new members for the sake of new blood is self-defeating. If their importance is greater than that of the existing community, we'll lose our experts in our mad rush to place the new user experience over that of our established (and valued) members. That's who this site is built around: expert gamers.
This site isn't a forum, where all you need is to register a new username and away you go. We have rules and policies that we follow, built on the history of StackExchange, and Arqade's own. Yes, it's a barrier to entry. Yes, we should be trying harder to welcome new users and teach them the ropes.
But we shouldn't be doing it at the expense of site quality. If a new user can't be bothered to read the rules that pop up the first time they ask (or answer) a question, or worse, read them and ignore them, then their question deserves to get closed if it's problematic. It's not something we should be waiting to do, just because they're a new user. Closure isn't about the user; it's about the question. If a new user wants to take it personally and leave, that's their choice.
TL;DR: Compromising site quality for the sake of new users is infinitely worse than closing their questions. Especially when it's a temporary state.