Just as a reminder the original Super User contest had four awards:
- The best rookie performance of a new user this week, as measured by the Super User leagues, will receive a 32 GB USB key.
- The highest reputation produced this week, as measured by the Super User leagues, will receive a 22″ LCD monitor.
- The “most awesome” new Super User question or answer that week, that most embodies the type of Q&A that make the site great — as judged by the Super User community moderators — will receive a two bay NAS device.
- The most useful Meta Super User question or answer of the week — as judged by the Super User community moderators — gets a Super User t-shirt and stickers.
While I certainly think this can be expanded on and improved, I'm not sure I entirely agree with Ivo that "most reputation by new user / any user in this week" is a bad contest criteria:
because of the highest reputation produced in that week in some cases resulted in 'negative activity'. Users would simply post any answer, see if it stuck and else deleted it again. Clearly not the kind of activity we strive for
Based on the 4 weeks of the contest, there was one user who was clearly gaming the contest. But the rest of the winners were quite legitimate and quite a few of them turned into avid SU community members. Judge for yourself:
- Monday, August 30th — winners announced
- Monday, September 6th — winners announced
- Monday, September 13th — winners announced
- Monday, September 20th — winners announced
So based on 1 out of 16 data points, I think it'd be a very bad idea to discard 15 excellent results.
(Also, even at worst, the user who gamed the contest did produce reasonable posts during the contest -- and didn't disappear completelydidn't disappear completely. So the 1/16 bad result isn't that bad.)