# What is the record for the fastest close vote, and which question achieved this accolade?

What is the record for the fastest close vote, and which question achieved this accolade?

I wondered if somebody has access to this data and I was curious about the distribution around the shorter end of the scale.

It is a shame this question was not closed faster... I was hoping to do the double.

• There's plenty of offensive and pure crap stuff that I recall to have been closed within the grace period. Sometimes even deleted too. – Asaf Karagila Jun 29 '18 at 14:42
• @AsafKaragila I think there are also examples which fall into neither of those categories which come nowhere near that upper bound. – samerivertwice Jun 29 '18 at 14:49
• Maybe. I think that your question here is bound to get closed and/or deleted. I don't think it's a productive question, and I don't know how someone without access to deleted-questions-data dumps can even answer, unless they happen to remember something very specific. – Asaf Karagila Jun 29 '18 at 14:51
• @AsafKaragila aah maybe you're right. I think sometimes it happens a bit too fast I guess. But in other people's eyes I guess even that isn't fast enough. – samerivertwice Jun 29 '18 at 15:02
• Are you wondering whether any of your own questions are in the running, Robert? – Gerry Myerson Jun 30 '18 at 4:51
• @GerryMyerson there was one recently, which was long and could not have even been scanned in 15s and had a close vote it seemed in around that time. It may have been justified but yes, it did make me wonder if I had achieved a record. – samerivertwice Jun 30 '18 at 6:42
• I find the closure reason rather strange here. I think this is absolutely an on-topic question that is clearly about site statistics. Whether it's an interesting or useful question is far more dubious, but I wonder how much of the closure is voting on the asker and not the post. – user296602 Jun 30 '18 at 15:44
• @T.Bongers So I can ask, separately, each to a meta post: What was the most quickly accepted answer?; What question received the quickest upvote compared to other questions in with low-upvote-times? What answer received the quickest upvote ever? What answer was deleted (other than self-deleted) most quickly? etc. etc. Let me know when I can begin flooding meta with such nonsense? Where do you draw the line? In response to Gerry's question, the OP admitted the question was motivated by a "questionable" belief that they were the victim of the most quickly downvoted/closed question ever! – amWhy Jun 30 '18 at 19:11
• @amWhy So I can ask, separately, each to a meta post: What was the most quickly accepted answer?; What question received the quickest upvote compared to other questions in with low-upvote-times? What answer received the quickest upvote ever? What answer was deleted (other than self-deleted) most quickly? etc Local context aside but, yes, those would all count as valid meta questions IMHO. – dxiv Jul 1 '18 at 4:48
• @dxiv: I'd agree (if something seems wrong with the spirit of the question, it might be better to have an answer dealing with that), but with the latest edit the question is now plainly teasing. I'd like to see that edit rolled back when the question gets reopened. – ccorn Jul 1 '18 at 11:17
• @ccorn I'm just making light of the conflict that happens here sometimes. I'm not against the crusade to eliminate unwanted content but I do think it's too fervent sometimes. – samerivertwice Jul 2 '18 at 13:10
• @RobertFrost: I understand that, but such things get old soon. Besides, as the distribution shows, there is a long slow tail; I surmise that this question simply has not been bad enough to be within those 5% that get closed quickly. IMHO there is no cup to earn here, only malus points for loss in style if that edit remains in place. – ccorn Jul 2 '18 at 19:44

The fastest close vote is hard to tell (without developer-level access to the Stack Exchange database). SEDE stores close votes but the timestamps are truncated to 12am UTC the day there were cast.

We can however check how fast questions are closed, provided that they're not already deleted by e.g. Roomba. This query does that; as of now, these are the top results:

Note that the first question has already been deleted; SEDE is refreshed once a week, on Sunday morning.

• Wow. 15s is pretty quick for complete closure! But I find it reassuring that it was a duplicate which can at least be reliably identified relatively quickly. – samerivertwice Jun 29 '18 at 17:58
• In fact it's really reassuring that virtually every entry is a duplicate question. However I also note the top 15 is dominated by 2017-18 entries, suggesting we either close more now or the system has got faster at allowing us to do so. – samerivertwice Jun 29 '18 at 18:02
• I believe you can blame Roomba for that. – Glorfindel Jun 29 '18 at 18:03
• FYI, the first one is closed by a mod and only the mod, so it is not that five users voted to closed. @RobertFrost – user99914 Jun 29 '18 at 18:14
• @RobertFrost the top 15 is dominated by 2017-18 entries Coincidentally or not approach0.xyz was announced in the fall of 2016, which made locating duplicates a lot easier. – dxiv Jun 29 '18 at 23:10

As a sort of "see also", I'd like to add that I have recently created some queries to investigate the cumulative distribution of time-to-close or time-to-deletion:

Update: Replaced a min with a max in the SQL (because ntile counts from 1, and I want to use 50 as marking the median). Updated results accordingly.

E.g. about 5% of the closed questions found in SEDE got closed within 20 minutes. Median time-to-close is about 6.7h however. When considering deletions as well, about 10% got closed/deleted within 20 minutes, Median time being 10.1h.

Caveat: My SQL is rusty, and it is easy to misinterpret data. When experimenting with SEDE, I often end up with lots of variants of one query. Better craft your own variants according to your curiosity.