Writing this has convinced me 'is drawing too many bad answers that aren't getting dealt with properly' is a close reason we really should add to our custom close reason list. Thoughts?

I remember reading another recent meta where a good question had to be closed because of this as well, and the poster was civil about it and everything. I think it would make people feel better if we could discriminate between closures on the basis of answer quality and closures on the basis of observable question quality.

Example time! (feel free to add more if you have any, metas preferred. I'm trying to go newest-->oldest):

https://rpg.meta.stackexchange.com/a/6947 (assume the question wasn't unclear anymore)




  • \$\begingroup\$ That's your user ID. Getting the link from the [share] button automatically adds that, so that the site can give you credit for any incoming visitors who followed your shared link posted outside SE. (It can be safely removed, too, if credit is irrelevant/unwanted.) \$\endgroup\$ Commented Apr 14, 2017 at 19:48
  • \$\begingroup\$ This is usually what Protecting a question is for \$\endgroup\$
    – Brian J
    Commented Apr 20, 2017 at 15:25
  • \$\begingroup\$ @BrianJ That definitely helps, but if the problem is with mid-to-high rep users it isn't sufficient. \$\endgroup\$ Commented Apr 20, 2017 at 18:02

2 Answers 2



A question that receives multiple bad answers shows that something in the question needs clarification or tinkering. If the question were in an answerable state, to the standards of the Q&A nature of the stack, then it wouldn't generate bad answers except for duplicates, unsubstantiated claims, or spam. Each of those can be dealt with by the voting system, flagging, and protecting the question.

Additionally, if we were to hold questions and provide a reason of "bad answer generation", that doesn't give the querent any feedback on how they can update their question to get it open again. So then we have to have a long discussion in chat/comments about the direction that the querent needs to take their update, which will likely end up being one of the standard hold reasons: too broad, opinion based, or unclear. So it's less work and less confusing for everybody to just use those hold reasons from the beginning

  • \$\begingroup\$ It is true that this doesn't tell the OP to change their question. That's the point. We've had a handful of cases recently and many more in the past where a question with nothing wrong with it is closed as opinion based because it's drawing a primarily opinion. If the only argument that can be made for a question's opinion-based nature is that it's drawing bad answers, I think we need to reconsider that position. Sometimes the answer isn't 'edit the question to be less opinion based', sometimes its 'there's nothing you can do because there's nothing wrong on your end'. \$\endgroup\$ Commented Apr 14, 2017 at 17:19
  • 1
    \$\begingroup\$ Basically the missing link here is between a question drawing opinion and a question being opinion-based. Lets say someone asks "what book(s) contains the rules for psionics in D&D 3.5?" and there is a continual flood of 'answers' from medium-to-high rep users like "Psionics sucks" "Psionics is the best" "My Dm made a fun homebrew" "Who cares about the rules, just make something up" etc. As soon as they are deleted, more are posted. Closing that question as opinion-based is... kinda unfair. \$\endgroup\$ Commented Apr 14, 2017 at 17:24
  • 1
    \$\begingroup\$ @thedarkwanderer If there are so many examples of this happening, you shouldn't use hypotheticals to back up your point. Use a real example. The kind of interaction you suppose wouldn't be solved by closing the question, it would be solved by using other systems like flags, deletes, and post protection. What you describe I think, respectfully, does not describe reality \$\endgroup\$
    – Adam
    Commented Apr 14, 2017 at 17:28
  • \$\begingroup\$ @thedarkwanderer Yeah, the responsibility for people being off-topic in the answers won't be laid at the question's feet. We're all smart enough to know that's not the causal relationship. (In that hypothetical example, it's clear that the causal relationship is the subject—psionics—and people's uncontrollable urge to slag them, not anything wrong with the question. Yes, that would be unfair to close over. We have better tools for that situation.) I begin to get the feeling that you're assuming a rigid decision flow chart that says “bad answers” → “close PoB”; that's really not the process. \$\endgroup\$ Commented Apr 14, 2017 at 19:51
  • \$\begingroup\$ @SevenSidedDie What? Why would that be unfair to close over? I think that sort of question absolutely should be closed, I just don't like that our only current options mislead the asker into thinking they did something wrong. If a question is drawing a flood of bad answers it really should be closed. We can open it again later after we figure out why the community suddenly decided it's okay to post answers like that. \$\endgroup\$ Commented Apr 14, 2017 at 20:14
  • \$\begingroup\$ @thedarkwanderer No, that's not how closing works. Closing is for when there's something wrong with the question. Most of the time we can tell by reading the question that something's wrong with it. Sometimes it's not clear, and we have to apply a litmus test to find out. When it's abundantly clear that a question is fine, we won't close it just because there are bozo answers. We'll just delete those answers. Here in this meta Q, abundantly clearly fine questions aren't at issue, only the ones that are borderline and need the “is it drawing crap?” litmus test to resolve the uncertainty. \$\endgroup\$ Commented Apr 14, 2017 at 20:19


There are two criteria being evaluated in this discussion: quality of the question, and quality of the answers.

For the sake of discussion, let's say quality can either be high or low. For this discussion, the querent is putting forth: what if there are high quality questions that receive low quality answers?

The thinking is that if a question is low quality, it will always attract low quality answers. Whereas, if a question attracts low quality answers, the question must be low quality as well. Well... not necessarily. This doesn't always happen in practice.

For an example of a well-formed question (as decided by the community), take the following: What level should this exhaustion-causing spell be?. While it is open now, it was closed twice and re-opened twice.

There was a meta discussion regarding this in which it has been expressed that this question is still too broad and opinion-based, even after the 2nd re-opening. Because of the apparent lack of great answers and visibility, a Bounty was also placed on this question by Ceribia.

The top answer for the linked meta discussion states the following:

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.

It seems, then, that the closing of the question was because it pulled in bad answers. But the quality of the question itself is not actually poor. A support of this is (1) the decision of the community to keep the question open, and (2) the similarity of that question to this and this -- similar questions which were not closed for pulling in high quality answers.

In other words: it can be shown that in reality, sometimes, a good question does get closed by no fault of the querent's, but simply because it has attracted low quality answers.

As such, we should be able to close questions with the reason that the answers it has pulled are of low quality, and not because of the question being of poor quality itself.

Another answer in the linked meta discussion brings up the alternative option of aggressive answer deletion instead of closure of otherwise good questions. This can be considered as well, and while it seems like it might be controversial, it does seem like a good solution for, say, answers with very low scores due to a large amount of down votes.

  • \$\begingroup\$ I think the missing link here is that the issue isn't about high-quality questions being deemed actually low-quality. Holding isn't for quality, up/down-votes are. So a question can be high-quality while still being held as a poor fit for the SE Q&A format. The thesis is not that attracting low-quality answers means the question is low-quality, it means that it has a not-ready-to-be-answered-here problem — possibly a problem not fully visible until magnified by the types of answers it draws. \$\endgroup\$ Commented Apr 14, 2017 at 17:03

You must log in to answer this question.

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