Inspired by "How often are closed questions re-opened?" on MSO, I wanted to find out how often questions on Programmers get re-opened.

One of the answers there provided two queries to get fairly close to the answer. Query One and Query Two brought back 8783 closed questions and 743 reopened questions.

That gives Programmers a ~8.5% reopen rate for closed questions.

However, the super sneaky and amazing Shog9♦ used his super-duper developer powers and came up with much more detailed numbers since his developer access gives him more details than what us mere mortals can glean from data explorer.

So, inquiring minds want to know. What's the real reopen rate for Programmers?

Bootnote: Shog specifically asked me to open this question here on MPSE

  • The query linked by VoteToClose for deleted posts is incorrect and can't be used for accurate reopen rates. The closest users can get is a guess, based on a combination of successful close/reopen counts in Data.SE (links to queries found here), and by guessing the deletion rate based on watching how many posts get deleted each week (by the roomba and otherwise) with this query. Or in 10K tools. – Rachel Aug 3 '14 at 16:16
  • Hey @Rachel, good to see you dropping by. Thanks for poking into those queries. I know you've spent some time trying to tease apart the answer to this question so I believe your analysis of them. My hope is that Shog9 will be able to run the same queries for Programmers that he did for StackOverflow. That 8.5% seemed high, so I was already a little mistrustful of those queries. – GlenH7 Aug 4 '14 at 2:03
  • 8.5% is way too high, just checking on the 10K tools every so often will tell you that. For example, we've had 11 questions reopened in the last 30 days. The list of closed questions caps out at 45, and the latest one about 30 hours ago (that list is supposed to show the top 50, so it sounds like 5 recently closed questions are deleted). If this were kept consistent, it would put us at 1,200 closed questions in the last 30 days, or < 1% reopen rate. I know that # is wrong because you can't avg a month on one day's data, but track it for a while like I did and you'll see its not that far off. – Rachel Aug 4 '14 at 17:45

First, the all-time stats:

Closed Closed then Edited Closed then Reopened Closed then Edited then Reopened 
------ ------------------ -------------------- -------------------------------- 
34567  4388               1174                 545                              

So, slightly over 3%, jumping to 12% for edited questions. For the past 365 days, we see:

Closed Closed then Edited Closed then Reopened Closed then Edited then Reopened 
------ ------------------ -------------------- -------------------------------- 
11824  1084               263                  140                              

Or roughly 2%, jumping to just shy of 13% again for edited questions.

The more interesting breakdown by close reason follows, again looking only at questions closed within the past 365 days:

Name                       Closed Closed then Edited Closed then Reopened Closed then Edited then Reopened 
-------------------------- ------ ------------------ -------------------- -------------------------------- 
duplicate                  713    49                 22                   10                               
off-topic                  6257   406                103                  48                               
primarily opinion-based    1598   187                53                   29                               
too broad                  1761   185                33                   15                               
unclear what you're asking 1494   257                52                   38                               

Expressed as percentages:

Name                       Closed then Edited Closed then Reopened Closed then Edited then Reopened 
-------------------------- ------------------ -------------------- -------------------------------- 
duplicate                  6.9 %              3.1 %                20.4 %                           
off-topic                  6.5 %              1.6 %                11.8 %                           
primarily opinion-based    11.7 %             3.3 %                15.5 %                           
too broad                  10.5 %             1.9 %                8.1 %                            
unclear what you're asking 17.2 %             3.5 %                14.8 %                           

This is mostly in keeping with what we observed on Stack Overflow, but it's worth noting that both duplicates and opinion-based questions stand a slightly better chance of being reopened, particularly if they're edited.

  • 3
    Many thanks for running the numbers and assembling the statistics! – GlenH7 Aug 4 '14 at 23:59
  • 1
    Does this data include Deleted requests? Also, Data.SE shows 5649 open questions in the last 365 days, which puts Programmers at a 67% closed/deleted rate. Is this correct? – Rachel Aug 5 '14 at 6:03
  • 3
    This includes questions that were closed in the past year even if they were later deleted, @rachel. That ratio is a bit off though, as I'm ignoring questions that were deleted without being closed, and also those closed but not asked during this time period. It's close enough though - the actual % is just a hair lower. Fwiw, you can get that number via search: programmers.stackexchange.com/… – Shog9 Aug 5 '14 at 6:13
  • Thanks for the clarification Shog :) – Rachel Aug 5 '14 at 14:05

You must log in to answer this question.

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