What level should this exhaustion-causing spell be? was asked and includes an element of homebrew, but has only 11 specific answers it can possibly be.

The question presents a given spell, and asks which level it should be. The possible answers here, given the 5e system, are: cantrip, 1st level through 9th level, and no level (ie, the spell is unacceptable).

On further checking, justifications that answer givers have provided have a fair bit of opinion, but seem to all be based on sound reasoning and experience. There was even a frame challenge to update the spell and make it better.

To contrast, I have asked two questions along the same lines before, and neither were closed. One was of a homebrew metamagic asking for how many Sorcery points it should cost, of which there would only be a specific numerical answer; and the other of a homebrew magic item asking for the appropriate rarity rating, of which there can only be five answers (Common through Legendary).

Though the question about spell level is about homebrew, it seems that it can be answered based on comparing the effects of this to pre-existing spells and conditions. For instance, the given spell was clearly better than a cantrip or a 1st level spell. There are guiding principles at play that experts can draw on.

I suppose, bringing up the existence of the homebrew tag also justifies this a bit.

Should the question have been closed, and why? And the more general question here is, should all questions of "[How Many/What Level/etc] of X should this Homebrew thing have" be closed as well?


3 Answers 3


The problem with this question is not "how many answers there are", it's that answering it has a too heavily subjective component. There are "too subjective" yes/no questions as well, the set of answers is completely irrelevant.

The spell is very unusual. Has anyone playtested a spell like this before? If so, they'd have standing to answer. But all the answers and comments-in-answers are pure guesses. The answers to your questions were more tightly scoped and got more objective content (though the accepted answer to the second one notes that this might be opiniony in its first sentence).

This is an example of a question being declared subjective because it's pulling subjective answers. In some other world where it didn't, it would be fine. But the assumption is that if a question is pulling subjective answers there's something wrong with the question. So it gets closed to retool.

So no, "all questions about homebrew categorization" should not be closed. Neither should they all remain open. They should be evaluated on their own merits.

If you don't believe that question should be closed in its current form, vote to reopen. If 4 people agree with you, it'll get reopened. If not, help it get fixed.


mxyzplk's answer is generally correct, but I prefer aggressive answer deletion to closure of otherwise good questions. I'm voting to reopen this question, and if you wanna help by voting to reopen as well and downvoting existing insufficiently supported answers en masse, that'd be cool too.

Mass downvoting seems superior to both closure and forced deletion in many ways-- if we can reverse the trend in voting on insufficiently supported answers to these questions we wouldn't need to close anything just for 'attracting bad answers'-- we could protect such questions from new users who don't get it yet by using the protect feature, while the rest of the site would learn not to post such answers because they will be scored negatively. Most of our users avoid behaviors that result in significantly negatively scored answers, as a general policy, even if those behaviors actually still generate reputation (like, if I get two upvotes and end up with an answer at -7 total, I've gained 2 rep. Nonetheless, I'll likely delete that answer and even if I don't it's grayed out and at the bottom of the page, below even new answers, and no longer a problem). Not sure we can actually change voting trends like that, but hey, it's probably worth a try.

  • \$\begingroup\$ Although, actually, only one of four answers is unsupported... \$\endgroup\$ Apr 1, 2017 at 20:05
  • 3
    \$\begingroup\$ Strong downvoting of insufficiently supported answers is definitely superior to closing a question. It tends to be the latter happens because our community isn't doing the former. \$\endgroup\$ Apr 1, 2017 at 20:22
  • \$\begingroup\$ @doppelgreener absolutely, which is why I think it makes sense to encourage the former. If we do a good job, then we won't need the latter ^^ Although in this case most of the answers seem fine. They're using a rules-based analysis rather than results from playtesting, but those are both valid ways to support stuff. \$\endgroup\$ Apr 1, 2017 at 20:59
  • 1
    \$\begingroup\$ Funnily enough, this question itself seems to be worth a second look. It has actually been re-opened already, and now re-closed. I'm not sure how many times it's happened (minimum of once I supposed), but the community seems to be very confused about what to do with this question. \$\endgroup\$
    – user27327
    Apr 2, 2017 at 2:35
  • 2
    \$\begingroup\$ @markovchain well, I've voted to re-open it (again), but re-opening is much harder than closing so the closers will almost certainly win soon. It is exciting to see reopening actually happen by votes more than once, though! (assuming this second re-open push happens). This is actually one of the rare times a bounty could be strategically used. If it gets reopened, you can quickly place a 50 rep bounty to keep it open for the next 24 hours, guaranteed. By then, the close-voters will have suffered the same time-based attrition situation as the re-open voters. \$\endgroup\$ Apr 2, 2017 at 7:10
  • \$\begingroup\$ @thedarkwanderer Seems like question has been re-opened and a bounty has been placed. :) \$\endgroup\$
    – user27327
    Apr 3, 2017 at 6:52
  • \$\begingroup\$ Bounties used to defeat the QA purposes of votes can easily be removed&refunded by mods, keep in mind. If we want to see this promote good behaviour, a line should be drawn at using bad behaviour to support it. \$\endgroup\$ Apr 3, 2017 at 7:21
  • 1
    \$\begingroup\$ @SevenSidedDie Using bountied to keep a question open is not bad behaviour. Bounties keep questions open for a reason. If you really think this really needs to be closed so much you'll play the mod card then sure, but wanton use of said abilities would strike me as bad moderation practices. Furthermore I would hope you would take the fact people are willing to bounty a question open after it's been reopened several times as indication that the topicality of this question is super interesting and important and worth discussing separately on meta rather than just making a descision. \$\endgroup\$ Apr 3, 2017 at 8:10
  • \$\begingroup\$ I mean, I haven't seen this kind of flopping on a question before, and I think it's indactive that this is really, really a great question to discuss for determining our scope because it's really very much on the edge of something. I think that in such cases it's important for all of us, especially high-rep users, to decide what to do about whatever underlying issue is at play. I, uh, didn't realize the question would turn out to be this big a deal when I posted this answer, I kinda want a new meta now ^^;;;;; \$\endgroup\$ Apr 3, 2017 at 8:12
  • 2
    \$\begingroup\$ Flopping is by design, and should be allowed to proceed, if it's going to proceed. It means the community is still deciding, and emergent decisions on main are at least as important as what talking on meta indicates. Bounties keep questions open as a necessary side effect. Using bounties only for their side effect — specifically to foil voters until it's off their radar — isn't okay though, because it interferes with our direct indicator of community position. \$\endgroup\$ Apr 3, 2017 at 14:10
  • 1
    \$\begingroup\$ As an analogy, editing posts bumps their question and moves other questions down the front page, for good reasons. Making many strategic edits to posts specifically to push another question down the front page would be using editing for ulterior purposes, and would swiftly trigger intervention. \$\endgroup\$ Apr 3, 2017 at 15:37
  • \$\begingroup\$ @SevenSidedDie isn't one of the okay reasons for bounty-ing to get a question attention from answerers? It would seem like any time you're willing to spend rep on a bounty to keep a question answerable that reason would necessarily apply. I agree it does subvert community process, though, but it can only be used once the question is more than 24 hours old and open-at-present. In any case, I agree it's reasonable to overrule it in the same way it's reasonable to overrule edit-bumping: when there's clear abuse. \$\endgroup\$ Apr 3, 2017 at 16:29
  • \$\begingroup\$ That's a reason to bounty, yes. In this case though, it's a dubious argument that it's the primary, innocent reason because 1) it clearly has garnered a lot of attention, 2) interfering with voting has already been suggested as the reason to bounty it at all, and mods are not expected to ignore what we've seen. \$\endgroup\$ Apr 3, 2017 at 16:55
  • \$\begingroup\$ @SevenSidedDie Oh, no, I was saying that interfering with voting is okay because that seems to imply that that's an okay reason to interfere with voting. I totally agree the real goal here is to interfere with voting. I agree that bounties used to interfere with voting outside their allowed uses should be stopped, it just seems like with that reason allowed the allowed uses includes 'cause I want this question open'. \$\endgroup\$ Apr 3, 2017 at 17:08
  • \$\begingroup\$ That would be one of the many corner cases where the system is not designed to foil misuse because it would be super-complicated to solve that odd corner case. In other words, the site shouldn't be interpreted as “strict RAW lets me do it, so it's okay.” Mods are by-design included because all the corner cases can't be handled by software rules. \$\endgroup\$ Apr 3, 2017 at 17:10

I did provide an answer to this question and the one thing that's really irking me about the question is the overall format we have on this site. Specifically, that the querent picks the 'right' answer without any vetting on their part. I personally feel strongly that the spell should be at least 7th level, but an answer suggesting that it's 5th level was chosen as 'right' and I don't think it is.

To be fair, no one can know for sure without at least some playtesting and I'm not sure how we negotiate that issue. Regardless, as we cannot compel someone to playtest to ensure that it works, I'm sort of leaning towards closure of the question if for no reason other than avoiding the appearance that there is a 'right' answer when none of us have done the necessary due diligence.

I don't think that necessarily all homebrew questions should be closed. As was mentioned previously, I also think that each needs to be evaluated upon its own merits.

  • \$\begingroup\$ I feel like closing a question because you disagree with the accepted answer is not in the spirit of closing questions. But I did come to the same conclusion as you, using a different method for analyzing the spell. \$\endgroup\$
    – user27327
    Apr 3, 2017 at 14:58
  • 2
    \$\begingroup\$ Keep in mind that the checkmark means only which answer was most helpful to the asker, nothing more, and nothing about correctness — it means “accepted”. This is by design, and is why you can see a -7 answer marked “accepted” beside multiple 10+ answers. If you object to an answer, as usual you can downvote it. \$\endgroup\$ Apr 3, 2017 at 15:34
  • \$\begingroup\$ @markovchain My issue had less to do with my disagreement over the accepted answer and more to do with I wasn't sure that there was a valid way to evaluate it. However, if the checkmark is simply indicating that the querent found that answer most helpful as SSD stated, then I suppose that's a valid metric. I just know we often close things as opinion-based when there's no way to reach a single 'correct' answer. It's a mantra I see repeated often, so I presumed it to be the best way to evaluate things. \$\endgroup\$ Apr 3, 2017 at 16:14
  • 2
    \$\begingroup\$ This is indeed why the checkmark is a very separate mechanism from voting. The checkmark is only what best helps the answer (and it's up to them, not us, to say what best helped them), but the community can feel free to assess the accepted answer as being no good or vote other answers as more appropriate. (This separation is also why SE staff have consistently declined feature requests to let the community override a querent's usage of the checkmark & force a different answer to be checkmarked.) \$\endgroup\$ Apr 3, 2017 at 16:16
  • 1
    \$\begingroup\$ I disagree that no one can know for sure without playtesting. This is a game design question and, while 'you must playtest as much as possible; playtesting is the best and only way' is a very common mantra in game design with RPGs, you don't just make an RPG out of playtesting. Analysis of the rules and their interactions, of third-party material discussing what you're doing, analysis of similar situations in play and test cases, all of these can be used as very strong support for an answer. Answers can be judged on the basis of the quality of their support. \$\endgroup\$ Apr 3, 2017 at 16:36
  • 4
    \$\begingroup\$ Since there is indeed no valid way of evaluating this spell short of playtesting, it should have been closed as "opinion based." But everyone wanted it open, so it's open, and this is the problem with that. \$\endgroup\$
    – mxyzplk
    Apr 3, 2017 at 16:51

You must log in to answer this question.