Existing Guidance

Reviewing guidance on meta and our post notice on main, it seems we put forward as a matter of substantive policy that answers must be backed up and supported, either objectively with proper citation to other relevant sources, or subjectively through a helpful description of your experience implementing your answer at the gaming table. Indeed, this is the guidance put forward on our canonical question on the issue, What are the citation expectations of answers on RPG Stack Exchange? :

People ask about mechanics because they're trying to better understand the game and how it works. Anyone can make a ruling, but that includes the person who asked the question, so we're not looking for this content. You should answer with how the game affirmatively handles the situation based on its text and cite the game's text to back up your statements.


Any of us can say “here's what I'd do” based on no actual experience, or come up with something on the spot as an off-the-cuff idea, but our site is not looking for this content. We want to collect tried-and-tested solutions with well-understood outcomes. We don't want your opinion; we want your expertise. If you do not have experiences you can bring to the table in that particular case, do not answer the question. Answers not doing so may be downvoted and/or deleted.

Now, this is guidance from meta, and the line between meta guidance and policy is often unclear (see here and here). That said, it seems to me that guidance is heading toward being policy when steps are taken to enforce it on main, such as our post notice, which reads:

Want to improve this post? Provide detailed answers to this question, including citations and an explanation of why your answer is correct. Answers without enough detail may be edited or deleted.

Meta guidance and the post notice seem to be pretty clear about this: unsupported answers are subject to deletion. Obviously, we should leave comments and work to help OP improve their answer, but ultimately, it is OP's responsibility to provide the meaningful justification for their answer; and if the answer is deleted, they can always come back to it and edit it, then flag for undeletion.


That said, are an answer and its author excused from this requirement if the answer is well received by the community?

Consider this question: What is the precedent for what happens to a warlock's character levels if their patron dies?

Obviously, the easy way out is "It's up to the DM". What I'm actually looking for is if there has ever been a precedent set for what happens when a character basically loses access to a class. For instance, 5e Oathbreaker Paladin shows what happens to a paladin that breaks their oath.

I'm interested in finding any precedent set throughout the history of D&D that will help a DM decide how to handle this.

The author sets a very clear expectation of what constitutes an answer to this question. This is an objective question; OP is looking for relevant source material that addresses the situation in question.

On this question, the most well received answer (by a significant margin, answer is at +83/-2, accepted and next highest answer is at +46/-4) fails to cite any of its claims. The answer reads, in full:

A warlock is defined by a pact with an otherworldly being.

Why would they think that killing the patron ends the pact? Sadly for your warlock, the obligation comes from the pact, not the patron. No doubt the dead patron has heirs and assignees who will explain this.

That’s assuming that the pact hasn’t already been restructured into several CPOs (Collateralized Pact Obligations) and on-sold on the infernal markets which have recovered well since the Multiverse Financial Crisis a few years back.

Just an alternate thought.

The answer makes many claims that are directly relevant to OP's situation, but not a single one of them is backed up by any relevant lore source. Normally, "here's a neat idea" answers like this are quickly downvoted and swept under the rug through normal community moderation. But this didn't happen here. Based on the existing guidance found here on meta, it seemed that a comment would be appropriate, encouraging the author to provide citations. The comment was responded to by OP including a meaningless link to the Warlock class description. I flagged for mod attention, suggesting that a post notice be added. No post notice was added, and further flags were declined with this message:

This is not a piece of curation override the mod team will be doing. If you disagree with the answer, please use your normal tools.

My normal tools here would be flagging LQP, which cannot be done, flagging NAA, which was declined by a moderator, linking in chat suggesting downvotes so that the answer could eventually be delete voted, which I'm sure 82 downvotes is out of the question, and flagging for mod attention to handle this strange exception to typical voting patterns and site curation, which was declined by a moderator, asking me to use my normal tools.

The only remaining tool is a meta discussion about if we really mean that unsupported answers are subject to deletion, or if totally unsupported answers get a free pass if they reach a critical mass where normal site curation tools cannot be applied.

  • 2
    \$\begingroup\$ I think the lack of citations is an issue for a lot of answers, but int his case, I think the bigger issue is that it's just Not An Answer. It's a funny comment, and those are okay - but I don't think they belong in answers. \$\endgroup\$
    – NotArch
    Commented Mar 12, 2021 at 15:58
  • \$\begingroup\$ @NautArch If the claims made in the answer were backed up by a proper source from established lore I think it would be a good answer, at least, part of it would be. The part about CPOs is probably totally irrelevant. \$\endgroup\$ Commented Mar 12, 2021 at 16:00
  • \$\begingroup\$ Let's assume for the sake of discussion that the answer is "No." What is your desired result, then? \$\endgroup\$
    – Novak
    Commented Mar 13, 2021 at 0:07
  • \$\begingroup\$ @Novak “Meta guidance and the post notice seem to be pretty clear about this: unsupported answers are subject to deletion. Obviously, we should leave comments and work to help OP improve their answer, but ultimately, it is OP's responsibility to provide the meaningful justification for their answer; and if the answer is deleted, they can always come back to it and edit it, then flag for undeletion.” \$\endgroup\$ Commented Mar 13, 2021 at 0:30
  • 4
    \$\begingroup\$ Yes, I saw that in the text of the main question. But "subject to deletion," is your phrase. The actual phrasing in the post notice is "may be edited or deleted." So again, using the Warlock question as an example, please tell me: If the answer to your question is, "No," exactly what do you want to see happen? \$\endgroup\$
    – Novak
    Commented Mar 13, 2021 at 10:03

3 Answers 3


Deleting a well-received answer is an act that should be reserved for extreme abuse.

Not because they are “excused,” but because those votes represent the “will of the people,” and no one here has the authority to overrule that. In short, they aren’t excused—they are simply deemed acceptable by the community.

Ultimately, “policy” here exists in the form of consensus. It’s not for the moderators to decide; it’s up to the community to decide. And when you have a highly-upvoted answer, the community has indicated its consensus that the answer is a good one.

It is not the place of any one user—moderator or otherwise—to countermand that consensus. Honestly, it’s not even Meta’s place to do so—the main site has vastly more participation than the meta site. 83 votes—up or down—is basically unheard of here.

To conclude, being highly-upvoted doesn’t mean an answer is immune to our expectations—it means that, in the consensus opinion of the site, it meets them. If you disagree, that’s fine—downvote, and/or start a bounty seeking better answers or pointing to a better one. You could even call attention to it in Chat, or in Meta if you absolutely must, to see if there are more who share your opinion and will also downvote. But those are your options. It’s not your place—even if you have a diamond—to override the consensus of the community. And the consensus here seems to be that you are wrong, it does meet our expectations.

  • 2
    \$\begingroup\$ Are you saying it is impossible for community voting patterns to be inconsistent with community established expectations? \$\endgroup\$ Commented Mar 12, 2021 at 17:23
  • 4
    \$\begingroup\$ @ThomasMarkov SE is not trying to create a hive mind. When I used that term I got a pile of approbrium from a couple of diamond mods. the general term is "wisdom of the crowds" and yes, crowds can sometimes make odd choices but here's the thing: if they find the answer useful why is someone complaining about the answer when they themselves don't find it useful? I am with KRyan on this. \$\endgroup\$ Commented Mar 12, 2021 at 17:29
  • 4
    \$\begingroup\$ Our tools are 'down votes' and 'comments' to alert the answerer and any reader to a need for improvement. Deleting stuff for being abusive, on the other hand, is a special case. An outlier. And as always Write A Better Answer is a brilliant response to a poor answer, or one rendered irrelevant by a variety of factors. \$\endgroup\$ Commented Mar 12, 2021 at 17:31
  • \$\begingroup\$ @KorvinStarmast Is current meta guidance concerning unsupported answers inconsistent with what you think best practice is? \$\endgroup\$ Commented Mar 12, 2021 at 17:37
  • \$\begingroup\$ @ThomasMarkov And you may also want to recall that comments that are funny and get a lot of flags are generally Left In Place and not removed. If you want to know where a meta post like this comes from, I am going to lay a twenty dollar bet in Vegas that it's the thought process behind this meta post that evokes that response. Can't say for sure, as I can't read minds, but that's a guess. Your objection to that answer ignores its value to more readers than you. They don't all see through your lens, nor mine. \$\endgroup\$ Commented Mar 12, 2021 at 17:40
  • \$\begingroup\$ @ThomasMarkov Which takes me back to - "hive mind" is not the goal of RPGSE. In the time you spent to complain about that answer, you could have written a better answer. (and given your talent, it likely would have been an excellent answer). \$\endgroup\$ Commented Mar 12, 2021 at 17:41
  • \$\begingroup\$ @KorvinStarmast I'm not sure a post that seems like it was made in bad faith is a good idea to use as an example. Whomever wrote that hasn't returned since posting it. \$\endgroup\$
    – NotArch
    Commented Mar 12, 2021 at 17:44
  • 2
    \$\begingroup\$ @KorvinStarmast We've arrived here because I am trying to consistently apply guidance from meta. Does the existing guidance concerning unsupported answers need to be hedged for responses that appear inconsistent with community established guidelines but are well received? Presently, I'm not aware that any of the community guidance suggest we should take into account voting distribution when judging answers, but this answer and your comments seem to suggest that we should. \$\endgroup\$ Commented Mar 12, 2021 at 17:45
  • 6
    \$\begingroup\$ @ThomasMarkov “We've arrived here because I am trying to consistently apply guidance from meta.” Well stop—that’s not what Meta is for. I am not saying that inconsistencies can’t happen—I’m saying inconsistencies are inevitable, and that’s OK. We aren’t trying to produce a perfect checklist for establishing what is or isn’t a valid question or answer; we’re trying to discuss some of the more challenging situations that maybe don’t have an immediately-obvious “correct response.” That’s it. But ultimately you have the options before you, and responsibility to apply your own judgment. \$\endgroup\$
    – KRyan
    Commented Mar 12, 2021 at 17:59
  • 2
    \$\begingroup\$ @ThomasMarkov Ultimately, in this case, there is no mechanism where the “will of the people” here could or should be overridden—the community has decided to endorse this answer. I don’t much like it, and clearly neither do you, but it’s not our place to decide for everyone. It’s not the moderators’ place, either—they didn’t sign up for that responsibility, and neither SE nor the community gave them that authority. In theory, Meta could have it, but you’d never get a Meta consensus of 83+ people, which makes any action Meta took here highly suspect. So we should just leave it. \$\endgroup\$
    – KRyan
    Commented Mar 12, 2021 at 18:02
  • 2
    \$\begingroup\$ @KRyan I do feel like it needs to be said that when Mods do act and override the community, it is generally as a 'community member' and not a mod. When that happens, the line becomes blurred in my opinion and brings up this type of moderation as possible. But that's probably another discussion to happen somewhere else. Just saying that the site, the users, and the mods aren't perfect and this question brings up voting as popularity contest vs voting for good answers. \$\endgroup\$
    – NotArch
    Commented Mar 12, 2021 at 18:06
  • \$\begingroup\$ @NautArch I mean, it’s the classic case where the correct answer to the question is “No,” and people don’t like that and refuse to leave it at that. This is not a new problem; Stack Exchange can and will always have cases like that. You don’t want to close questions just because the answer happens to “No,” (not least because the querent probably didn’t know that), but when it happens this is often what you get. HNQ also plays a role. But none of that makes it a good idea to try and undemocratically overrule the votes. \$\endgroup\$
    – KRyan
    Commented Mar 12, 2021 at 18:10
  • \$\begingroup\$ Thank you for engaging respectfully and openly! Do you think we should continue to nip answers like this in the bud, but it's more that once flowered, the moment is gone? \$\endgroup\$
    – NotArch
    Commented Mar 12, 2021 at 18:17
  • \$\begingroup\$ @NautArch I think we “should” more-or-less exercise our rights and responsibilities (as indicated by our privileges and the buttons put in front of us) as we see fit. The “delete” option isn’t one of those buttons on such a highly-upvoted answer (it is for diamond mods but IMO not for this purpose), but would be for a downvoted answer, so “early” could make a difference. (I’m not sure that’s a good thing, though; the inertial tendencies on voting are a bug, not a feature, to my mind.) \$\endgroup\$
    – KRyan
    Commented Mar 12, 2021 at 18:52
  • 1
    \$\begingroup\$ @NautArch Also, here, I’m not sure I would vote to delete this even if it were heavily downvoted—it does at least hint at the answer to the question. “Why would you think that” implicitly suggests “there is no evidence of that,” and ultimately, that is the correct answer. It’s not clearly stated, and then the answer goes off on tangents, but there is a kernel there that answers the question (and correctly). So I would downvote, but not vote to delete, myself. While we can delete answers for “very low quality,” the hint at the correct answer here is enough for me to say it isn’t that. \$\endgroup\$
    – KRyan
    Commented Mar 12, 2021 at 18:55

No, it doesn't eliminate standards

I'll start with a direct answer to the question as asked: no, a high score doesn't remove the need to adhere to community standards. The particular answer linked is not a good one, and your assessment of it as a comment (which would also not meet community standards) seems correct to me.

The most important issue with it is that it directly ignores the criteria laid out in the question. Had the question not asked specifically for precedent throughout all of published D&D, it might possibly have squeaked by some way or another. But it is clearly not providing what the querent requested and is not a frame challenge. Matters are complicated very slightly due to the querent approving of the post but that doesn't change any of the specifics.

It is probably worth noting that, while the answer is deficient in its lack of citations or explanations of rules, the answer isn't quite wrong-- its content agrees with the accepted answer in that nothing happens to the Warlock's class levels. This is a not-too-unusual occurrence of the frequently encountered case of the rules not containing specific information. You can't cite rules that don't exist.

It also features a literalist interpretation of the Warlock class (it's defined by the pact, not the entity the pact is made with). That could have been expressed more clearly and fully in that answer, but could be read as affirmative evidence explaining why there is no precedent for anything happening: because the pact endures, nothing happens.

I don't love the answer, but I don't think it's as fundamentally transgressive of site standards as is asserted here. That doesn't mean that deletion is off the table, only that it may not be absolutely required.

When you don't get the result you want after exhausting the available tools, you don't get the result you want

If this is the least satisfying experience you ever have on RPG.SE, I envy you! Thinking that X should happen, according to how you interpret site policy and purpose, and X doesn't happen, that's very frustrating. I think that we've all been there at some point. But there is no closely held principle here that you (or any one of us) eventually end up precisely where you'd like to be in all matters. You've used most of the tools available, and once you've used all of them then the conclusion of those efforts is the conclusion even if you don't love that end state.

You have not exhausted your normal tools

This does not address the general case of "amusing" content being left up while not quite adhering to site standards. But, given that the answer to this question appears to be "nothing happens to the Warlock class levels", you could edit the answer pretty easily to suit:

There is no official mechanism regarding what happens to those class levels when the patron dies, and so there is no particular reason to think that they will be lost. I'm not aware of any instances where that loss occurs in game lore. The power gained from being a Warlock may not even be from the patron, exactly, but could be interpreted as being derived from the pact itself. That could outlast the patron, for example by [the rest of the answer as written].

That's a pretty hacky kludge, but I hope the idea comes across. As there does not seem to be a citation to use, a sentence to that effect transforms the rest of the answer into a more palatable one (with regard to adhering to standards, at least).

The goal of the site, broadly, is to produce good and useful answers. Deletion is sometimes a part of that, but the best general impulse is probably towards improvement. Even if a user hates a particular answer, the right first tool to grab is not a sledgehammer.

You have, or almost have, your answer anyways

You've gone through most of the normal processes, and have not seen the action you think is most appropriate. As above, I (like many users here) am familiar with that frustration. But it is, if not quite over, nearly so. You've outlined the steps you took to try and cause deletion of the answer, and the answer was not deleted. The flags were declined, mods declined to curate to your preference, a campaign to rally users in chat to downvote the post into oblivion didn't pan out. Now there is this (and possibly another, forthcoming) meta discussion which seem like an extension of that campaign, though at the time I'm writing this the community does not seem to be solidly in agreement with you here. If these tools don't work to secure your preferred result, they don't work. There are reasons that users, even high-activity users doing a lot of community mod work, do not have a "delete this answer unilaterally" button.

By all means, use the tools that are available. If you think that the mods are mistaken, you can open discussion (as was done here on meta), send a complaint to SE staff, run for mod yourself, and so on. You are entitled to the available tools, but not so much to your preferred outcome(s).

Rules for the sake of rules are not really what the site is about

The answer is not great, but I don't see that it does much to damage the site or the experience of people coming to use it. Obviously a large number of users thought that the answer was good enough to deserve upvotes, and community opinion is the core of SE overall. "I have net-ten upvotes on a meta question" is a poor basis to fundamentally override that. Further, there is not a strong consensus that pluralities of votes on meta create binding policies which require direct enforcement by everyone in all cases-- rather the opposite, in fact.

A nontrivial part of why community action is so quick on questions and answers that are deficient is to avoid this exact scenario: poor-quality content which becomes popular is a different animal than poor-quality content that virtually no one has seen.

I think that it's generally good to promote and apply best practices on the stack because they improve the stack: better questions, easier-to-find questions, clear answers, citations supporting those answers, descriptions of experiences in practice-- all of these, in aggregate, tend to make the site more useful. That's good! It's what we want, and helps make the site more useful overall. But, really, the stack is virtually identical with this answer remaining up as it would be with the answer deleted.

Looking at old posts and taking the strongest possible action based on the appearance of technical violations of loosely-held preferences about standards as voted on by a tiny minority of stack users on meta is not the hinge on which the entire stack turns.

tl;dr: We can and should enforce "back it up" standards for answers, and I believe that we are right to look poorly on "try this"- and "I imagine this to be true with no support"-type answers. Those are true even if this particular answer falls through the cracks: it's not great, but not necessarily, obviously a flagrant violation of a core site need. But this site is about users more than rules, and if there is not sufficient user support to commit to an action it may not happen. That seems to be where we are here.

Working to promote best practices in future posts is virtually always a better way to support the stack than trying to force the past into present compliance.

  • \$\begingroup\$ Great answer.for the specific example. \$\endgroup\$
    – Akixkisu
    Commented Mar 16, 2021 at 18:47
  • 1
    \$\begingroup\$ Very Nice Answer \$\endgroup\$ Commented Mar 18, 2021 at 17:23

We should use the tools that we have.

We want to both respect the votes, that answers accumulate, of onlookers and community members but also the policy that our community establishes over time.

We do this by raising a flag "In need of moderator intervention" and write down that the answer in question needs a banner that informs the reader that answers that aren't backed-up don't meet our standards.

Moderators then will act accordingly in their role as janitors and add the banner:

Want to improve this post? Add citations from reputable sources by editing the post. Posts with unsourced content may be edited or deleted.

  • \$\begingroup\$ In this case moderators refused to add the post notice and responded that they would not be taking action on this post. I explained in the second to last paragraph of the question that I had already exhausted all the tools available, besides meta discussion. \$\endgroup\$ Commented Mar 12, 2021 at 19:08
  • \$\begingroup\$ @ThomasMarkov you raise a meta post. You showcase that the post doesn't meet our standards, as one of the answers. The moderator in question (or somebody else) will respond there with an opposing answer, and the community can decide if the banner is warranted or not by regular vote. \$\endgroup\$
    – Akixkisu
    Commented Mar 12, 2021 at 19:15
  • 1
    \$\begingroup\$ Isn't that sort of where we are? I've exhausted the tools available and raised a meta post. \$\endgroup\$ Commented Mar 12, 2021 at 19:16
  • \$\begingroup\$ @ThomasMarkov Yes, and no. You raised a meta post about the practice. Next, you ask a Meta about the specific answer - in a Q&A style, your answer will argue the case to use the banner. Then we proceed. \$\endgroup\$
    – Akixkisu
    Commented Mar 12, 2021 at 19:25
  • \$\begingroup\$ @ThomasMarkov what you did here was reviewing what kind of policy we have and showcasing your specific exxample. But what you want to do is to vote about a specific action. \$\endgroup\$
    – Akixkisu
    Commented Mar 12, 2021 at 19:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .