I think that's a good thing to do, but it doesn't specifically address the issue I raised in the other topic about "contest bias."
The problem I see is that, when participation in a contest isn't high, or when the quality of the entries isn't great to begin with, even the winning entry might not be very good. The quality of the result of the contest can be no better than the quality of the entries. So, I thought it might be a good idea to give the community some sort of "veto power" after the winner is announced as a failsafe, by having everyone vote "yes" or "no" on whether the winner should be submitted to Greg for canonization. Moreover, a second poll would further validate the claim that, yes, this is a MOC that the community is generally satisfied with as a canon representation of the character and does not offend too many people.
I'll give you an example, which hopefully will convince you that this is a problem worth considering. Suppose you had a contest with four entries, A, B, C, and D, and the results were 26% of votes for A, 24% for B, 25% for C, and 25% for D. A is technically the winner, but only narrowly. Should it be made canon? In the extreme case, up to 74% of voters hate it or at least prefer not to see it made canon. Or, at the other extreme, maybe all of those 74% of voters would be fine with it being canon, but had a different first choice. The purpose of a second poll would be to determine which case you are dealing with. Maybe the votes were split because none of the entries were very good. Maybe it was because all of them were super awesome. You wouldn't want to canonize the result in the former case, but you would in the latter case. The second poll would be an impartial and fair way of determining which it is.
This problem might very well not be an issue today, though. I just wanted some form of quality control to avoid marginal contest results like, say, these ones: