# Serial downvoting

I received 4 downvotes within a short amount of time, I suspect it has been the same person.

I think a mechanism should be installed that no user $u$ can downvote more than one post with a positive score of another user $v$.

It's not dramatic, but it feels like "downvote trolling", and I think we should try to avoid this on mathoverflow.net

• If I'm not mistaken, the SE software has a variety of mechanisms in place to detect abuse and correct for it; you may well find that these downvotes are reverted once the software does its job. – Carlo Beenakker Apr 7 '18 at 19:00
• To add to @CarloBeenakker's comment, links to some basic info about this can be found in serial-voting tag-info. However, only 3 of the downvotes are very close to each other, I am not sure whether that is enough for the reversal script to kick in. – Martin Sleziak Apr 7 '18 at 22:55
• Thanks for your comments! Maybe I should delete this post then if the information is otherwise available? – Dominic van der Zypen Apr 8 '18 at 6:06
• Dominic, let me take this opportunity to express my appreciation for all your activities on MathOverflow. I've enjoyed so many of your questions, which I find so often to be imaginative, playful and interesting. To my way of thinking, one learns mathematics best simply by playing around with ideas, combining them and seeing where they lead, and my impression is that you follow a similar philosophy. Please don't let a few downvotes get you down. Most of us recognize that people who ask many interesting questions are the main source of success for MathOverflow. So keep up the good work! – Joel David Hamkins Apr 9 '18 at 13:05
• Many thanks, @JoelDavidHamkins - and vice versa, I am often amazed by your deep and original answers to all sorts of questions! It escapes me how you can be so incredibly productive. I look forward to many more answers (and questions) of yours! – Dominic van der Zypen Apr 9 '18 at 13:20

I understand that getting downvotes can be frustrating. And targeting a specific user with multiple downvotes is definitely a bad thing to do and it can in fact lead to a suspension of the serial downvoter.

But there already are mechanisms that try to catch such instances automatically. For instance, in this specific case you can see on the OPs reputation tab reputation change +6 on April 8 2018 with the description: "Voting corrected (learn more)".

In the cases where the automated script does not revert the downvotes even after some time has passed and you still suspect that you have been targeted, you can flag for moderators' attention. Moderators have some additional tools to investigate and they can also ask Stack Exchange staff for help.

Having said that, I disagree with the proposed feature request:

I think a mechanism should be installed that no user $u$ can downvote more than one post with a positive score of another user $v$.

I think that there are situations where downvotes are legitimate. Implementing your suggestion (or something similar) would restrict possibility to downvote. And it strengthens the asymmetry between downvotes and upvotes. (There already is some asymmetry in that downvotes have lesser effect on the reputation of the poster than upvotes. Moreover, for a downvote on an answer the downvoter has to "pay" one reputation point. And only users with repuation above 125 can vote down, while only 15 reputation points are required to be allowed to vote up.)

In fact, this might have the opposite effect and lead some users to be more prone to downvoting, after thinking along these lines: "Well, this post is borderline, some users might upvote it, some users might downvote it, some might even vote to close or delete. I should better downvote it quickly - if somebody else upvotes it, it might be possible that I am no longer allowed to downvote."

And the criterion as described in your post is rather unclear. This question is a good example. If you look at the timeline then you can see how the score of this question evolved. Namely the post went from $0$ to $-2$ (two downvotes on April 7), then to $+1$ (after three upvotes on April 8) and it is currently back at zero.

If you suggest that system should check downvotes on the questions with positive score, how would this question count? After the score flipped from negative to positive, what would happen to the downvotes cast before that?

• I think your criticism of my proposed mechanism is valid and therefore accepted this answer. Thanks for posting it! – Dominic van der Zypen Apr 16 '18 at 9:05