This post is an extension of a related thread on meta: Does your decision to upvote/downvote depend on the current score?

The above discussion mostly focuses on the phenomenon of upvoting questions and answers that one feels have been unfairly downvoted, or received insufficient attention, even if one might not have upvoted them purely on the posts' own merit.

However, I am interested in the reverse behavior: one sees a very popular post which seems to be pretty mediocre, but is not an intrinsically flawed question or answer - it just doesn't seem to merit its current popular status. Is it acceptable to downvote in such a situation?

To give a concrete example, let's say I am browsing the top-voted questions in the [geometry] tag looking for interesting problems to ponder, and come across this post. I might think:

The question is somewhat well-defined, but it's not all that interesting, and really quite easy to answer with a graphing calculator. I don't want to see this sort of thing when looking for captivating geometry problems, and I would have been happier if it had been lower in the queue so I could have instead seen better questions. I think MSE would be a slightly better place if it didn't implicitly present this question as the second-best out of 40,000 geometry problems on the whole website. And the user hasn't done anything else on MSE in the past decade, so I'm not likely to hurt their feelings.

Nowhere in this thought process is a judgment that the question is actively bad, or that it would merit a downvote if it was sitting at $0$ in the new queue, but it does present a case for downvoting the post as it currently is, so as to bring its score in line with what it "should" be.

In this particular case, the question is closed and so the point is moot, but it attracted $180$ downvotes before closure, so evidently this kind of thought process does happen rather often on the site.

The case against this sort of thing, as I see it:

The help page says "voting down a post signals [...] that the post contains wrong information, is poorly researched, or fails to communicate information". If a post does not have any of these flaws, it should not be downvoted. Your votes should be a function of the merit of a post, and not a tug-of-war against other users' enjoyment of it.

The case for:

Math StackExchange should serve as a lasting resource for many future users to view posts relevant to their interests. If you are willing to spend a few reputation points of your own in order to improve the relative position of relevant questions and answers, thereby making the site a slightly better experience for everyone, this is a valuable service to provide and should be encouraged. Post scores should be a function of their merit, not of their view count.

What is the community opinion on such behavior? Is this sort of action acceptable? Should it be accompanied by an explanation, if so?

  • 10
    $\begingroup$ If you think it is okay to upvote a question that you deem too negatively received, then I'd think you'd think it's okay for someone to downvote a question with a score out of sync with its merit. $\endgroup$
    – amWhy
    Feb 21, 2021 at 18:10
  • 2
    $\begingroup$ @amWhy: Maybe so, but it seems like there is some asymmetry, both in how a user perceives being upvoted vs. downvoted and in the reputation cost to the voter themselves. I could see MSE cultural norms differing. $\endgroup$ Feb 21, 2021 at 18:13
  • 19
    $\begingroup$ Questions are illigiitmately upvoted for reasons unrelated to the quality of a question. A very poor post answered by 5 separate users may very well get five upvotes. Friends upvote friends, admire certain users, etc. To me those are abuses in voting far more often that retaliatory downvotes. $\endgroup$
    – amWhy
    Feb 21, 2021 at 18:17
  • 1
    $\begingroup$ People have a bias, all of us, to accept as true praise and upvotes because they make us feel good. People also have a bias, all of us, to get upset by, and question, critical feedback/downvotes. The asymmetry is that to the OP, they're biased regarding downvotes, are more than they have a bias regarding upvotes. So inflated upvotes, seems to me, needs ongoing scrutiny. $\endgroup$
    – amWhy
    Feb 21, 2021 at 18:21
  • 1
    $\begingroup$ ^^^ The above comment of mine is the asymmetry. SE sites are not the business of first and foremost, making everyone "feel good". I mean, the drug trade could argue "we feel an obligation to help people feel good!? But how valid is that. $\endgroup$
    – amWhy
    Feb 21, 2021 at 18:40
  • 1
    $\begingroup$ @RavenclawPrefect As amWhy points out perfectly , your post is a reflection of the asymmetric treatment that users give upvotes and downvotes. I'm not exempt! But over time, we must realize that our questions should be judged on merit. It's like a Markov process : forget about the history of the post, look at it as if it was just posted, judge it on merit and upvote/downvote. If this leads to a barrage of downvotes on a previously highly voted question, so be it. On the "for" part ... (continued) $\endgroup$ Feb 22, 2021 at 5:23
  • 1
    $\begingroup$ I personally think merit has to go above everything. Let me take your posts as an example : you explain your posts very well, honestly, but they contain very different content (tiling questions) to what we usually see on the site. However, if someone posted your question without giving any details, then sometimes , just because people are fascinated they up vote and start looking at it. That attitude has to be defenestrated. We must judge posts on their suitability to the site, and only that (in a "Markovian" fashion?). All other criteria come second, IMO. $\endgroup$ Feb 22, 2021 at 5:26
  • 2
    $\begingroup$ We are currently grappling with the issue of "poorly researched/presented" contest questions, which present the huge interest versus merit clash I spoke about earlier. With this in mind, my answer to the reverse situation will be : if it is not meritorious , then downvote. Leave a comment, if it is a relatively new user whose post has the behaviour, but please do downvote. The attitude of people towards downvotes, me included has to change. One way is to downvote constructively. For another post, but as amWhy says, the better way is to up vote constructively. +1 to your question, though! $\endgroup$ Feb 22, 2021 at 5:30
  • $\begingroup$ There are exceptions to the "Markovian rule" I speak of. For more content you may read here. $\endgroup$ Feb 22, 2021 at 5:36
  • 2
    $\begingroup$ I agree with the comments on asymmetry. I would also be cautious about downvoting a post which others have recognised without giving some thought as to why they may have appreciated it. There are all sorts of reasons, but it may just be that the post has merits for particular users. Mind you I have seen false and misleading posts given votes by people who haven't appreciated a subtle mistake - I have voted such down, but more importantly have made comments to highlight the mistake. [I have made such errors myself - it can be easy to go wrong - and I appreciate being corrected] $\endgroup$ Feb 22, 2021 at 19:39
  • 2
    $\begingroup$ On one hand, the merit of a question shouldn't depend on its current vote total; on the other, simple questions have much broader appeal and receive an inflated number of upvotes. The question you've linked doesn't bother me so much since it's from 10 years ago, when the site worked quite differently. But then there are answers like this one, consisting of two sentences with an extremely straightforward application of definitions, that somehow get 23 upvotes(!), while many more in-depth and insightful answers don't even receive 3 upvotes. $\endgroup$ Feb 24, 2021 at 18:40

2 Answers 2


From a rules standpoint, you can vote how you want, with only a few exceptions-- namely, revenge downvoting, targeted voting, upvote rings, etc. The common theme of these exceptions is that they're based on factors other than the merit of the post. In general, the goal is for your vote to be independent of extrinsic information like author and view count.

I would say that voting to say "this post is overrated" violates this principle. If you see a bad post that's been heavily upvoted, you should downvote it, but that's because it's a bad post, not because others have upvoted it. The opposite is true for good-yet-downvoted posts.

In practice, most people do upvote much more than they downvote. The extent to which this indicates an error is arguable. If math.se has mostly good questions, this is appropriate. Personally, I don't think this is true, and I'm inclined to say that people tend to be naturally downvote-averse. The optimal solution to that isn't to go out of your way to give "corrective" downvotes, but it may help to consciously remind yourself not to forget about downvoting as you use the site. In other words, the cure for inobjectivity is objectivity, not contrarianism.

As a final point, keep in mind that the most important votes to give when it comes to site health are for new, active questions, since this helps promote the good ones and throw out the bad ones. Downvotes are a good cue to check if something needs to be closed, if you're just scrolling through the front page. Voting does help rank older posts, but this is not as important.

  • 4
    $\begingroup$ AHA, like a true moderator! Indeed, people tend to be downvote averse. At least in my case, I broke the ice with this tendency only recently, and my thoughts were always : what if I wrote that answer/question which I were about to downvote, how would I feel? Not very good, right? Something like that , and then I would not downvote. But only recently have I started downvoting far more on merit, after seeing how it actually has an effect on improving the site. I think people need to see the improvement in site quality that constructive downvoting can bring about. The same goes for upvoting! $\endgroup$ Feb 22, 2021 at 10:01
  • 3
    $\begingroup$ Regarding people being too downvote-averse, is there a rule of thumb for what your upvote : downvote ratio should be? For me, I have upvoted $1000$ times and downvoted $50$ times; I'm unsure if this is too much or too little. $\endgroup$
    – Joe
    Feb 22, 2021 at 12:31
  • 1
    $\begingroup$ Freely recognising good contributions is helpful, I think. There are badges awarded for good questions and answers as recognised by votes, and that is part of the ecology of the site. I think these have become rather harder to come by as the site has got busier and the turnover of questions has become greater. $\endgroup$ Feb 22, 2021 at 19:33
  • $\begingroup$ I should add for new users' questions, it seems to me sensible to seek engagement and acknowledge any engagement you get, with some allowance for slower response times while the user works out how to respond on the site. If there seems to be potential for improving a question to MSE standards, I'd allow time for that to happen before reaching for downvotes or suggesting closure. $\endgroup$
    – Joffan
    Feb 25, 2021 at 18:35
  • 2
    $\begingroup$ @TeresaLisbon I for one believe downvoting should be done if and only if you are simultaneously adding a comment why. If you can't point out why you think the answer or question is bad, you can't possibly expect the person who wrote that answer (clearly believing it to be good) to somehow find out what you disliked about it. $\endgroup$
    – Poseidaan
    Feb 26, 2021 at 13:01
  • 2
    $\begingroup$ @MartinvanIJcken True. However, I usually don't do so if my message is the same as someone who has already commented. However, we have the same issue with up voting , which people don't think of as an issue. What do you like about my post that made you up vote it? I can understand the sentiments : for example, if I was grading a child and he/she got $10/10$, then what comment can I leave other than "well done!", whereas I'd have more to say if he/she got $4/10$. I still think that most up votes are unwarranted, and such behaviour as you mention should be encouraged for up votes as well. $\endgroup$ Feb 27, 2021 at 10:00
  • $\begingroup$ @MartinvanIJcken Furthermore, I am shifting my focus on the site from answering questions to moderation, at least slightly. Therefore, I am encountering the need to improve questions more. I now go through many more questions in a day and so I get to refine such bad practices as I had before. This includes putting your comment into practice, which I admit to be very efficient (like the practical evidence to your theory). If you'd like to know more about this, you can visit the CURED chatroom. $\endgroup$ Feb 27, 2021 at 10:23

Personally, it is difficult for me to base a downvote on whether the query broaches an interesting mathematical topic. The reason for my difficulty is that in my experience, about 95% of the queries that are posted are seriously defective.

By this I mean that totally ignoring the general interest of the mathematical topic, the query will typically have such problems as:

  • Original Poster (OP) not show work
  • OP not use MathJax
  • OP not provide the background of the problem (i.e. previous theorems or solved problems that lead up to the query that the OP thinks might be relevant)
  • OP omits their own level of math expertise, so it is unclear what type of analysis the OP will be able to understand.
  • Significant ambiguities/typos in the query, with no proof-reading of the query done by the OP.

Frankly if I see a query that does not have any of the above defects, it is like a breath of fresh air. In such a circumstance, I don't care if the OP is asking how to multiply $(7 \times 17)$. If the query is not defective, I am not downvoting it.


You must log in to answer this question.

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