This is my solution to the popularity contest problem.

  1. Have a meta post about making a good popularity contest.
  2. Popularity contests after being posted must set a time period of a minimum of one day before answers are allowed to be posted. Therefore there is not a luck element to them. This would allow people to make answers and would help remove fastest gun in the west.
    1. The other option would be to wait a minimum of one day then record the votes on all answers then the winner is the highest increase of votes.
  3. Popularity contests should be written with a Objective winning criterion. Think of creativity as the tool not the guide. Give a clear example of what a ideal answer looks like. Make sure to leave room for people to expand and have fun.
    1. Example of good winning criterion: "Voters should consider how well the program patches the text." It allows humans to judge how well the code did. A bad example is: "Vote on how nice it looks!" that is BAD everyone views it different. You can encourage making it look nice. Offer a bounty for objective things like that.
  4. Discourage new users from using it by saying (something to this effect) in the tag wiki snippet: "This tag is highly risky. New users should avoid using this tag without posting first in the sandbox."
  5. To quote a removed user: "Another thing you want to make sure of is that the challenge is still ontopic for the site. If there were a challenge "write code that draws a pretty image", popularity-contest would clearly be the best possible victory criterion. That doesn't make the challenge a good popularity contest, though, because it's a bad challenge, and giving it an appropriate victory condition won't change that. It's important to pick a topic for which writing a popular answer will take skill (especially programming skill), and where the subject matter isn't too far from the sort of subject matter normally covered on this site."

That is my personal view on what we should do. What do you think?

  • \$\begingroup\$ I posted as a new question as feedback is needed. If there is a better place to post this please tell me and I will move it \$\endgroup\$
    – user63187
    Jul 26, 2017 at 19:06
  • 4
    \$\begingroup\$ I suspect people are going to point out problems with your suggestions, but that's inevitable given that pop cons are a hard problem that may not have an ideal solution. I appreciate you trying to tackle this problem head-on and making specific suggestions. \$\endgroup\$
    – xnor
    Jul 26, 2017 at 19:18
  • 1
    \$\begingroup\$ I don't think 2 will really fix much, it will still be a FGITW just with a 1 day delay. I think the problems with popularity contests are too deeply rooted for us ever figure out a way to make voting representative of quality. We have to either accept that voting is based mostly on time submitted or kill the pop con. \$\endgroup\$
    – Wheat Wizard Mod
    Jul 26, 2017 at 19:19
  • \$\begingroup\$ 2.1. doesn't fix FGITW at all, because the problem isn't that early votes give you a headstart for the final tally but that early votes put your answer at the top of the challenge for anyone sorting by votes (probably most people, especially those arriving from HNQ). \$\endgroup\$ Jul 26, 2017 at 19:53
  • \$\begingroup\$ @xnor Thanks I am trying to help in any way I can \$\endgroup\$
    – user63187
    Jul 26, 2017 at 19:57

1 Answer 1


I do have posted and participated in a bunch of popularity contests and already discused their problems a lot, so I think I may adress some of the issues I've seen, what I agree on and what I do not agree on. I do recommend reading the previous discussions and decisions.

Regarding 1: Yes, all challenges should be sandboxed.

Regrading 2: If you have 1 day without answers, the challenge is going to get burried, and still not everyone can participate at the point of time you are allowed to post, so you just postopone the FGITW "problem". I think with the current format of this site there is no way to awoid FGITW effects or any other effects of votes that influence other votes. I just accept that now.

Regarding 3: The winning criterion is the number of votes by definition. We already have an objective validity criterion. And you cannot enforce a voting criterion, people are going to vote on whatever they like, so one should design a challenge where the goal is something that people will like.

Regarding 4: Yes that would be a good idea, but I don't think many users actually read the tag description as we already rewrote that a while a go to account for the points raised by many discussios before.