According to the canonical post on automatic bans:

Yes, ... deleted answers count towards an automatic ban.

Now, this is not very specific, and I suspect that is intentional. However there are certain cases where I don't think it would be fair for deletion to count against the answerer, and would love to see some clarification / guidance (without revealing any specific details about the algorithm):

  • answers that are deleted because the question was deleted. This was mentioned here, but no satisfactory answer yet. This typically will be through no fault of the answerer or any issues with the quality of their answer (though one could argue that a question that gets deleted should not have been answered - but then you get into whether the question was self-deleted, was it down-voted first, or did the community/moderators delete it).

  • answers that are self-deleted. For example, if I answer a post and then realize someone beat me to the punch with an identical answer, or perhaps later someone posts a better answer. This is peripheral to this recent discussion and again is not necessarily a quality issue with the answer or the author. In fact, we probably want to encourage, rather than punish, behavior that cleans up redundant or lower-quality answers.

In either case, I would hope that just the mere fact that an answer was deleted does not count against a user. In these scenarios it would be unfair, IMHO, unless the answer was down-voted or maybe flagged prior to deletion (meaning there was a problem with the answer, according to a user other than the author).

  • I think both bullet-points could reasonably somewhat count if they affect a high percentage of one's answers. If 90% of your answers are to questions that get deleted, you're obviously doing something wrong. Ditto if you delete 80% of your answers. But for reasonably small percentages of either ilk, those deleted answers shouldn't get you banned. – Daniel Fischer May 12 '13 at 17:38
  • @TinSoldiersAndNixonsComin' right, but my point is whether they should count at all. Since we don't know the algorithm (and I'm not asking for that knowledge), it is feasible that I could do this twice, and then post a bad answer, and get banned because I hit a theoretical threshold of three deleted answers - even though two of those should arguably not have counted toward the ban. In effect I could be answer-banned for one bad answer. Or to a greater extreme, 0 bad answers. – Aaron Bertrand May 12 '13 at 17:41
  • 2
    It's possible to interpret that statement as "deleted answers count just like undeleted answers". In other words, the deleted status has no weight in the measurement, but the votes do regardless of deletion. However, I have no idea whether that's the correct interpretation. – Josh Caswell May 12 '13 at 17:42
  • @Josh true, I hadn't considered that interpretation. If that is what it means, then I think the canonical answer should be updated to reflect a less ambiguous reality. – Aaron Bertrand May 12 '13 at 17:43
  • @AaronBertrand Yes, I'm not saying they should necessarily count against you [not YOU-you, of course], but something like if (percentage(deleted) >= n && count(deleted) > k) for sufficiently large values of n and k wouldn't strike me as unreasonable. – Daniel Fischer May 12 '13 at 17:49

You must log in to answer this question.

Browse other questions tagged .