1

The simple equation that decides whether you get question banned or not is:

(total questions - negative score questions - closed questions)/total questions >= 0.5

And closed questions probably includes deleted questions too. For example, a new user asks a question, and the question gets:

  • Downvoted, stays at negative score

  • Deleted before 30 days from posting time

  • Closed

The equation above would not work as:

(total questions - negative score questions - closed questions) < 0.5

Which means question ban. But that is way too early to get banned. It seems that the user has to reach a certain number of questions or rep or time for the above equation to be implied towards the user. I have a small grip on the subject myself. So when does the question ban equation get applied?

  • 4
    Where did you get that equation from? The question ban algorithm is a secret that is more closely guarded than the government's classified documents. – Mysticial Mar 25 '15 at 22:27
  • I found that on some post in Meta.SE. Is it that important and secretive? These users can't keep seem to keep that secret I guess. – Anthony Pham Mar 25 '15 at 22:27
  • 9
    Where exactly did you get these equations and figures? As far as I know the actual ones are not public knowledge - what you have there looks like a guess by someone. – Oded Mar 25 '15 at 22:28
  • I can't seem to remember about the user, but I do remember that the user who wrote this question had a lot of rep. Probably a trusted user – Anthony Pham Mar 25 '15 at 22:29
  • 13
    That looks like the formula for the Socratic badge family for maintaining a positive question record. Nothing to do with question blocks. – animuson Mar 25 '15 at 22:30
  • 11
    It's not that that formula is secret. It's that it's wrong. :) – Adam Lear Mar 25 '15 at 22:30
  • 1
    So should I delete this question? – Anthony Pham Mar 25 '15 at 22:31
  • I would like to know the approximate number at which you get banned. But I know the formula is secret, and what you have posted is for the badge, as said above. That's why I up voted @oded . – Tim Mar 25 '15 at 22:32
  • 1
    I think you can take the upvote as someone thinking that's the actual formula. Which it isn't. The premise of this question is completely wrong. As such, it doesn't make sense. – Oded Mar 25 '15 at 22:32
  • I don't know what happens faster, the comments or the views – Anthony Pham Mar 25 '15 at 22:33
  • 3
    @Tim We're not going to post numbers or reveal the specifics of the formula, sorry. There are a lot of variables that go into this, so there's no "approximate number" we could come up anyway. – Adam Lear Mar 25 '15 at 22:41
  • @Anna that's fair enough, better to keep this secret :) – Tim Mar 26 '15 at 9:27
14

The question ban is applied when a user who qualifies for it tries to ask a question. Until that point, they're like that cat in a box, both banned and unbanned.

For this to make sense, you must understand that your formula is... extremely inaccurate. As usual, I'm not going to publish the actual formula, but I'll make a few notes as to the nature of these bans:

  • There are multiple, cooperating algorithms at work. Some of them result in permanent bans, others just apply stricter rate-limits to askers.
  • Not all questions are taken into account. Everyone makes mistakes; banning you for one would be unfair... Unless making mistakes is all you do here.
  • Well-received questions count for a lot. Really, any participation that demonstrates you're not a complete drain on the system helps in avoiding the ban.
  • Zero-scored posts don't help at all. This isn't some MMO where you can grind your way to victory by mindless clicking; someone has to find your work useful.
  • The faster you're posting questions, the faster you'll get banned if those questions aren't well-received. Given it takes some amount of time for even well-written questions to attract attention, if you're hitting the rate-limits for questions per day/month and aren't also dripping with upvotes you should probably be somewhat concerned.
  • wonder if system somehow keeps the record of attempts to ask that trigger block (as far as I know, something like this happens to spam attempts) – gnat Apr 3 '15 at 9:45
  • 1
    Yes, it does @gnat. – Shog9 Apr 14 '15 at 7:30
  • 1
    Is there a grace period for new users / how strict we are with those who might not know how the community works?. Follow-up; Once you know the cat is dead, can you determine when it might come back to life? I've seen a question from someone who's answer-banned and wants to know if they've even got a chance of answering before actually spending time on writing an answer, which seems like a reasonable request. Say the threshold "ban score" to post is x and the user is on x-100, can you say for certain that no answer would be good enough? – Basic Mar 6 '16 at 4:02
  • 3
    Answer-bans are extremely weak, @Basic - they're meant to stop folks who are completely unable / unwilling to do ANYTHING constructive on a site. A nominal amount of positive participation is all it takes to get out of them. And yes, all bans have a grace period of some sort, and warnings. – Shog9 Mar 7 '16 at 22:39
  • Part serious, part joke: is there an emergency procedure in place if a disgruntled former employee publicly leaks the exact ban algorithm? – Sonic the Anonymous WizHog Feb 15 at 6:52
  • I suggest that it determines whether we are banned when we click ask question instead of when we click post – Asadefa Feb 15 at 18:20
  • It does, @Asadefa. Unless you're using the mobile app. Or the score of your posts change while you're writing the question. – Shog9 Feb 15 at 19:01
  • @Shog9, I beg to differ. I have experienced the annoying behavior of only finding out I have a Q-ban after I typed everything up and tried to click post. See meta.stackexchange.com/questions/310245/… – Asadefa Feb 17 at 19:50
  • The UI for "ask a question" and "answer" is rather different, @Asadefa. – Shog9 Feb 17 at 20:11

You must log in to answer this question.

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