Example Poll for possible future voting scheme

Terr

Cranky Coder
aa
Jul 31, 2009
1,590
410
Given some of the complaints people had in terms of the voting integrity/ease-of-use/etc. during the Swamp-detailing contest, I'm trying a voting system out (purely unofficially) and would like folks to vote in it.

This particular poll is entirely public (and thus, cheatable) but it has the option to send individualized "Ballot links" to a predefined set of "voters" by e-mail.

Simply rank things by how much you like them, the one you like most at the top, the one you like least at the bottom.

Click: Vote For Your Favorite Fruits

You can either rank everything start-to-end, or you can "group" things into your own tiers. Both are OK ways to vote if you really can't decide between certain sets of fruits. You can also select "no opinion" if you've never tried a fruit. (Admit it, you've had similar problems voting for maps, yes?)

The top three fruits will be ranked. My hope is that this may be a useful system to use for future contests. I'll close the poll later this week and post the results.

P.S.: The statistical math behind it which makes it mostly-fair is very interesting but entirely outside the scope of this post ;)
 
Last edited:

Lord Ned

L420: High Member
Feb 11, 2008
421
174
Voted. I find this way rather cumbersome. It might be better if the default answer was "No Opinion", because if I just wanna vote for myself or such, I'd hit "LOLMINE=1" and leave everything else at 18. Wouldn't that drag down those votes?
 

Terr

Cranky Coder
aa
Jul 31, 2009
1,590
410
if I just wanna vote for myself or such, I'd hit "LOLMINE=1" and leave everything else at 18. Wouldn't that drag down those votes?

Amazingly, no, it doesn't drag everybody else down. That's the beauty of it. :) If you rank one map at "#1" and rank 17 others are "18" that's just the same as ranking all the others at #2. It's based on your "this-or-that" decisions, rather than the numbers themselves.

Mathematically speaking, the winner is the one that would beat everybody else (with >50% of the vote) in every possible 1v1 contest.

[ame]http://en.wikipedia.org/wiki/Condorcet_method[/ame]
 
Last edited:

Terr

Cranky Coder
aa
Jul 31, 2009
1,590
410
OK, I remembered to end the poll ;P

Using the Condorcet-IRV method (which is resistant to "burying") the winners are Apple, then Strawberry, then Kiwi.
 

Armadillo of Doom

Group Founder, Lover of Pie
aa
Oct 25, 2007
949
1,228
I appreciate the effort Terr :) For the mini-contests, a popular vote was used b/c it's always been easier, and never had a lot of entries. For the major contests, if we get similar numbers, a popular vote will be used to trim down the entries, so the judges aren't trying to fully review 10+ maps in the space of a week.
 

Dr. ROCKZO

L8: Fancy Shmancy Member
Jul 25, 2009
580
159
Yes, with so many entries, picking a "number 1" is getting harder, giving the option to choose a list of your favourites would help and probably provide a more accurate opinion of the maps. You would have to be careful with how much you limit it though, if you made people vote for 10 maps, near the 8th or so map people would just be clicking on random names to fill the quota.

Chances are I've completely missed the point, or stated the obvious, but that's my 2 cents.
 

Tapp

L10: Glamorous Member
Jan 26, 2009
776
215
I think that right now one could simply duplicate the electoral system. Post a poll, wait for votes, and eliminate the loser (s). Repeat. While it would require more polls, it allows for optimum opinionating. [/my 2 cents]
 

Terr

Cranky Coder
aa
Jul 31, 2009
1,590
410
It also allows for shinnanegans where people "bury" a candidate by voting for a choice they have even less intention of supporting in the final round.