This is gonna be long. You've asked some very important, very complex questions and I'll try to do them justice--but you don't have to take my word for it. As we get deeper into the material I'll be offering links to references, histories, and further explanations of various themes and topics. I highly recommend using those links as jumping-off points for your own research.
First:
You're absolutely right that sometimes folks are overzealous about curating the RPG Stack. People get excited or anxious or just plain crabby, or a particular topic strikes a little close to home. We can misread a post, or misunderstand a policy, or just be so thrilled about a new tool that we overuse it for a while. This behaviour needs to be called out --kindly!-- so we can improve both as individuals and as a community.
What happened here specifically?
The examples you've invoked are, to my eye, edge cases which can swing either way. There is no “close questions exactly like this” memo you didn’t get, people are just making judgement calls from experience-based site principles. Our community usually makes reasoned calls and is open to reasoned challenges--I've successfully argued for reversals myself, but arguing a specific vote reversal is different from challenging a general principle. Trying to do both at once usually accomplishes neither.
As the community and its policies grow and change, we don't do active witch-hunts for older questions no longer consistent with new policy or understandings of what works here. Older material was handled in ways out of sync with how we currently operate. One question was asked two and a half years ago and the other was asked yesterday; different members of the community responded to each according to the era. Questions like yours are often how older questions get modern curation applied to them!
Mxy's comment explains what new learning was applied to the newer question: experience shows that even if a question seems superficially fine, serially poor answers indicate there's some underlying problem which needs addressing. (You may not feel the answers are poor; that's something I'll talk about downstream.)
Closing policies in GS/BS: A Big Topic
Your broader question is about closing policies in general and GS/BS in particular, and that's a really big topic! I've taken years to feel like I've wrapped my head around it. [deep breath] Okay, here goes.
The first thing I had to figure out is that each Stack site is a dynamic organism. They've all got the same skeletal structure of site mechanics, and the same broad directives and guidance from the network, but it's a living process to learn how a particular Stack's community and subject matter best weaves those elements into a good Stack site for their topic. RPG.SE is always learning more about itself and shifting its behaviour accordingly. What it looks like today is radically different from five years ago, and it'll be just as different five years from now. I expect this question will influence that future!
Early on RPG.SE was eager to accept every kind of question even vaguely related to our hobby; later on folks realised this wasn't always viable.
Also RPG.SE has learnt it's different from other Stacks because many RPG experts are used to traditional RPG forums. And, well, a lot of those forums have a reputation for caustic behaviour which is not so normalised in many of the other Stacks' wider related communities. Over the years this community has clung increasingly to the Stack network's central principles and guidelines --much more than some other Stack sites-- because it's sadly necessary. This isn't unique to RPG.SE, but you may not spend a lot of time in Stacks like Christianity.SE, parenting.SE, or gardening.SE (yes, really!) which have had it even worse than we do at some point and/or have had longer to figure out how to deal with it.
Moderation and curation are organic processes that need to adapt to emergent conditions. As Mxy points out, sometimes a bit of extra zealotry is necessary in the face of an influx of newcomers who are taking a while to process how this Stack is different from what they're used to. So when current policies get discussed, proposals for change should take into account both the history which brought us to this point and the current condition of the site. I'm glad you've asked about the background of these issues, and I hope this answer can help give some context for your concerns--and please do check out the links I'm providing, too! Don't just take my word for it.
You're absolutely right that the Stack can sometimes seem cavalier about our contributions to the site, and disdains fascinating opportunities related to a site's subject material. This can be very disheartening, but there are a few notions which may help explain why it happens and how it's often a feature rather than a bug (however bitter a feature it is to swallow). To that end, I'd like to take some space specifically addressing some of the mechanics and policies you alluded to in your question, and some references to RPG.SE's history with them. For whichever of these things you already totally get, just consider me nauseatingly thorough.
Subjectivity is awesome and we like it. However, over the years RPG.SE has learnt what the Stack network already knew: subjective answers need to be backed up with someone's relevant experience (not necessarily the answerer's own!). Unsupported speculation causes a lot of problems, from comment arguments to sorting difficulties to just plain useless answers. Besides, really good answers explain why and how their proposed solution is a useful one. GS/BS is just a codification of that principle. The Stack wants to draw on its users' expertise with whatever situation is being asked about, rather than collect a bunch of untested guesses that may not work or may even cause serious harm to games, interpersonal relationships, or people.
Up voting and down voting are wholly unrelated to close voting and to delete voting. I semi-regularly upvote questions which I also vote to close, because each kind of voting has separate unique criteria. The Stack network's goal is to provide well-sorted actionable answers to problems people are really facing, and up/down voting is (alongside tagging) how we do the "well-sorted" bit of the job. A well-upvoted question or answer should be useful (probably well-researched, clear, and interesting), but that doesn't make it any more or less in line with, eg, the site's topicality guidelines. It's also important to know that while there are guidelines for voting, everyone is free to vote as they see fit with no repercussions or oversight except for automatic checks against rampant vote-spam abuse.
- Closing is not a punishment, it's a reprieve. Closing a question gives the querent space to improve the question before they get inundated with answers that don't quite match what they were after (thus wasting everyone's time). It's also a firm flag to the querent that they should improve their question; we can't just hope somebody gets lucky and guesses a useful answer, it forces us to do our bit keeping up the site's content quality if we want solutions. Which leads me to the next bit...
The Stack optimises for answers, not questions. Questions are a dime a dozen and we can afford to be picky about 'em. The Stack wants good answers, and experience has slowly taught each site what sorts of qualities in questions attract more chaff than wheat. Please read the linked article; there's a cost/benefit analysis at work in the network which errs toward quality over quantity. You may have also seen some Stack sites where that priority is not given its due, often out of a desire to be welcoming at the cost of being useful. This causes problems because...
- Our questions define our answers. This is implicit in our advice on questions to avoid (admittedly a weird place to put this): "Constructive subjective questions: inspire answers that explain 'why' and 'how;' [...] tend to have long, not short, answers; [... and] invite sharing experiences over opinions." If all the answers on a particular question are poor in the same sort of way, we should look to the question to see why.
Deletion is not automatic, nor (with one exception) is it an escalation from being closed: it's a different mechanic entirely. (When questions are put on hold, they switch to "closed" after some days but it's just a cosmetic change and they aren't in danger of being deleted simply by virtue of having been closed.) It’s reserved for only the most egregious cases, usually spam, offensive content, or obviously non-Stackly answers. The stock notice about adding sources or the answer will be deleted is addressing those extreme cases so folks can't say they weren't warned: it's not a common or casual thing. (There was a period on RPG.SE when deletion was wielded more liberally for a very specific subset of poor answers to a very specific kind of question, but that era is past; it was a last-ditch effort to retain our ability to usefully answer game-recommendation questions.)
And now, some harsh and often unpopular parts of the Stack Exchange philosophy. These are the "love it or leave it" bits where a user can either live with it because the results are effective, or can't.
The Stack Exchange is not and never has been interested in being a one-stop-shop for all questions on a Stack site's target subject. There are many great forums which handle certain kinds of questions very well, and the Stack is mostly interested in filling some of the gaps which more discussion-y sites don't handle well.
Similarly, the Stack isn't interested in everyone's participation. Maintaining and growing a well-sorted stack of actionable answers to practical questions can be daunting, difficult, and requires cordial disagreement. Folks who aren't interested in doing that for free... well, we're okay with losing them quickly because it wastes less of their time.
Understanding what’s going on now
I'm gonna speak to your last two paragraphs directly now.
I agree totally that the community can be lax in explaining clearly why questions are being deleted, or why certain guidelines are enforced for answers. We struggle with this constantly: if we explain too much, it's dogpiling and lecturing. If we explain too little, it's esoteric and elitist. It's a fine line to tread and you can help us. Perhaps you could start by offering improvements to this sadly underdeveloped resource, or suggesting some new strategies entirely?
It can be unwelcoming when my hard-worked content goes away, even if I understand why. Our Stack network overlords believe quality content is more attractive to experts than warm fuzzies--though our overlords are happiest when warm fuzzies are also available to the users, that comes a distant second to content curation. We attract and keep active users more through making sure our site has lots of great content.
That said, the closed question you linked to will not be deleted, nor do I think any of its answers are particularly likely to be deleted. The existing content remains, sorted and tagged for everyone to benefit from. The closure was just a plug to so more users won't spend their time and energy on creating and curating unsupported speculation--which while interesting, doesn't attract and keep experts. Closures are easily reversed when the cause is addressed, and I hope this one gets addressed. It's only been a couple days now.
I hope you find RPG.SE palatable and continue to help us create and curate this excellent stack of solutions. If you do decide it's not your style, though, there's no harm nor foul. Not all Stack sites are equally appealing to me either; there are a couple with topics I love but whose communities and/or policies don't mesh with me at all. It's a little sad, but very understandable.
It's important for folks to give (respectful, constructive, supported) feedback on how each site can grow, so thank you for yours. I recently gave another Stack an earful, myself. I hope I've given you some context for what's going on here, and I look forward to your continued activity. Thanks again for asking those questions, it's been good for me to re-examine this stuff too!