Let us discuss the reasons that play a role when evaluating to vote for deleting a low-quality answer

Question

Let us discuss the reasons that play a role when evaluating to vote for deleting a low-quality answer.

Deleting answers perceived as low-quality is a contentious topic, and many cases aren't as clear-cut as we'd sometimes want.

I'd appreciate insights into various decision-making processes that different community members use for evaluating answers perceived as low-quality 1.) to reflect on my method, and 2.) to offer other community members to reflect on their process.

I don't intend this post as a prescriptive place for creating guidelines or best practices. Instead, I prefer it to be descriptive — focused on your concrete actions, not ideals.

I believe that edge cases where you walk us through your conflicting line of thought are most interesting.

Answer 1

The deletion privileges page reads:

You may vote to delete answers in the following cases:

• The answer is extremely low quality: There is little to no scope for improvement

Of course, this is extremely terse and doesn't provide much guidance. What does "scope for improvement" mean? Shog9 wrote an answer a while back giving his take on reviewing posts in the LQP queue, and I think it gives a helpful perspective:

I use a pretty simple algorithm when reviewing these, whether in /review or the flag queue:

1. Does the post attempt to answer the question? No Delete, Yes goto step 2.
2. Would it be possible for a reasonably-intelligent English-speaking person familiar with the topic to understand the solution being presented? No Delete, Yes Looks Good.

Note that I don't make any attempt to determine if the solution is correct or helpful or necessarily even relevant to the question as part of this process (if I do happen to be familiar with the topic, I'll go back and vote/comment on it later though). I definitely don't concern myself with the potential for author edits, unless the problems are such that I could fix them myself (in which case condition #2 prevents deletion). Folks whose worthless answers are deleted are free to come back and post new ones with clear, useful information if they care to.

My goal here is to remove the sorts of garbage that tend to infest popular forums after a while: questions as answers, idle kibitzing and gibberish/nonsense/stuff that's been through automated translation software a few too many times.

Also note that from /review/low-quality, delete / recommend deletion responses allow you add a predefined comment - these comments will alert the author of the post (with a link to it) even if the post is later deleted; they work essentially the same way as moderator comments in this regard, and for the same reason.

For me, this bit is a crucial expansion on the deletion privileges guidance:

I definitely don't concern myself with the potential for author edits, unless the problems are such that I could fix them myself (in which case condition #2 prevents deletion). Folks whose worthless answers are deleted are free to come back and post new ones with clear, useful information if they care to.

So one of my criteria looks something like "can I improve the post, or does it require the author to improve the post?" When I encounter a low quality, unsupported post that still appears to be relevant to the question, if I can improve it, I just do that, but if improving the quality of the post requires something only the author can provide, I'll lean towards voting to delete and/or flagging VLQ. The idea here is that if I can understand what the answer is trying to say, but maybe it just needs some citations, or the language needs to be clarified, the argument isn’t made as clear as it could be, or it’s missing pieces that I can fill in, I added those things with an edit; but if I can’t do those things, and can’t really see someone with more relevant expertise doing it, that’s when I lean toward a vote to delete.

To give an example of how I have applied this idea, consider this question: Is this homebrew Eldritch Invocation, Eyes of the Ooze, balanced? The question presents a homebrew feature and asks if it is balanced as written. The accepted answer is excellent: it provides analysis, compares to other features, and offers a suggestion for balancing it that doesn't involve drastic changes to the feature's function. However, this answer is not so great:

It needs a level requirement of 15 to make it balanced, but then it's pretty good. Still on the stronger side though.

That's the entire answer text; it provides a short opinion with no reasoning whatsoever. So reviewing Shog9's algorithm, yes it does attempt to answer the question, but what about the second criterion? While we can certainly understand that the answer's author thinks it would be balanced with a level requirement of 15, they have provided nothing to us to help understand why that is an appropriate solution, and this is where we ask "but can I (or someone other than the author with sufficient expertise) improve the post?" In this case, the answer is "no" - no one but the author can provide the author's reasoning. Sure, if you agreed that 15th level would be appropriate, you could come up with some reasons, but they would be your reasons, not the author's reasons. No one but the author can improve this post, so this answer falls under this bit of Shog's guidance:

Folks whose worthless answers are deleted are free to come back and post new ones with clear, useful information if they care to.

Answer 2

Two similar edge-cases that I still have no idea how to handle

There are two particular kinds of answers that I still have absolutely no idea how to vote on. Those being answers that are objectively incorrect and answers that are toxic and/or propose solutions that are deeply flawed and problematic.

For the first case, there are answers like this one which describes the notation "1d-4" objectively incorrectly, or this one which claims Rogues can only get Sneak Attack once per round. Both of these answers currently have a single delete vote.

---

For examples of deeply flawed answers, there's this one:

I hate to sound crass here, OP, but there's a reason that you don't date your fellow players, and you're discovering it. Love tends to make one see something in another that other people don't, just by its very nature. This means that even if you're best friends with someone, they might end up being in love with someone you despise, even though that seems bizarre. Not to mention people tend to treat games differently. Your friend's husband obviously doesn't understand the point of the game, or doesn't care.

The lack of perspective from the "innocent" member of the couple is what creates the rift in the table. You have to discuss this right now and as directly as possible before negative sentiments become more deep-seated and someone explodes. You will not get lucky and have your GM divorce her husband suddenly or something equally unlikely. The worst case scenario is that you leave the table quietly, but your friendship will remain intact, while a full-table drama episode could end up with everyone hating each other.

Beneath this answer is the following comment:

This attitude is condescending and toxic. Every couple I've ever gamed with has been terrified that their very existence could ruin the game, to the point that they impose pointless handicaps and limitations on their gaming habits. Couples have never been a problem in any of my groups, but this attitude toward them has. This querent's problem is a challenge faced by individuals, not an inevitable result of couples gaming.

And for another example, there's this one

[...] Handicap his character forcing him to develop his RP skills. This can be done a million different ways. ex: forced polymorph, death with new creation requirements, equipment loss, forced individual diplomacy, Stat losses, unusual poisoning/enchantment, charmed/cursed equipment, or just have the other players kill his character if he is doing "evil things" and causing the party harm.

There are a million thanks that would case a player to start questioning his actions. And if they complain let it be known, this isn't a video game. Your actions and or inactions have consequences.

Beneath this answer is the following comment:

I have intentionally not deleted this, as I believe (at time of commenting) -7 downvotes is more instructive than no answer. This is an abysmally wrong answer, however.

Many of the site's lowest scoring answers fall into one (or both) of these categories. Furthermore Meta Stack Exchange even has some guidance on this kind of stuff:

• Why shouldn't I delete wrong answers?
• If a valid answer is downvoted, what's the motivation not to delete it?
• Why is an answer that is so bad it is shown faded out kept around? Why not delete it?

We even have a relevant Meta question ourselves:

• Deleting of "bad" answers by Stackizens seems to be increasing

This has a few answers as well as this comment:

Sometimes keeping a bad answer is quite helpful: it's very instructive in how to not use the system.

---

I know there are people who think terrible answers, in at least some cases, should be kept around so that their highly-negative scores can showcase what not to do. But I've also seen a lot of answers that fit into one (or both) of these categories get deleted and I've seen a lot of arguments about how deleting these is actually the best thing to do. There's a lot of information on these with all sorts of proposed methods of handling them.

That all said, I don't expect this answer will be helpful for deciding whether an answer should or shouldn't be deleted. I wanted to talk about a particular part of that decision-making process that, even after a sizeable amount of time here, and even after having given this considerable thought, I still have no idea how to vote. People will disagree when it comes to these, and that's perfectly fine.

I'll end with probably my biggest takeaway from looking into these kinds of answers: deletion voting is not a science, it's not an exacting process with hyper-defined rules and objectivity; uncertainty is unavoidable and that's okay. Look around, ask around, figure out what other people think, and build up your own ideas about it. Trust me, there's a lot of discourse on this and just when you think you've found all seven ways to handle question X, you'll find somebody suggesting three completely new ones.