Inofficial proof of concept test poll (please participate)

Contests 4K Game Competition - 2006
#1

You can tick one, your top5, top10, top13… all… none… etc. Its so called approval voting. I think I’ll go with a top6 or top7 :wink:

Its supposed to be pretty fair and basically tactic free. (Of course its a bit skewed since we already know some possible outcome… but try to forget that completely.) So, I thought it would be nice to do a just for fun test run and if it works well we could use this voting system next year. And if it doesnt work… well, than we can at least exclude this voting scheme.

Poll details:

  • 55 options
  • a maximum of 55 votes per user
  • runs for 7 days
  • users are not allowed to change their vote
  • the results are shown after the poll has exipred
#2

I was very close to running this same test myself. Good idea, will help us determine a voting plan for next year. :slight_smile:

#3

I dunno; I have to say that I think “public” voting for future runs of Java4k isn’t a good idea- I suspect that it will open up the judging processes to a number of abuses (ballet-stuffing, etc) which the current judging system isn’t prone to. I worry also that having a voting/rating system would turn Java4k into more of a popularity contest than anything else.

Perhaps both a voting system and a group of independent judges could be used? I certainly wouldn’t object to having both a “popular choice” and “judges’ choice” category; I’m just wary of putting the contest up to public vote alone.

  • HC
#4

I understand your concerns and I also agree pretty much with your objections. The voting itself doesnt need to be all that public. Eg it could be restricted by filtering out users which signed up less than 6 months ago. Also if a lot of people vote it would be pretty hard to manipulate the poll w/o getting noticed.

And yea that “popular choice” and “judges’ choice” thingy would be pretty neat.

Well, either way… this poll is just a test for checking how usable the results are and how much they differ from the judges’ order. Hopefully it will offer some insight, which could lead to a more educated discussion about the voting scheme next year.

If the sample size is big enough we could also compare it to things like trimmed mean, median, or average of normalized points. Stuff like that. Just for the fun of it. :wink:

#5

Not at all enough votes to give a good result, I believe. Can be different with more votes :confused:

#6

Sorted results:

25 (17.1%) Miners4k 15 (10.3%) fuzetsu 15 (10.3%) Xero - The Last Hope for Mankind 7 ( 4.8%) Goomba4K 6 ( 4.1%) Roll 4K 5 ( 3.4%) Speed 4k 5 ( 3.4%) Icejump4k 5 ( 3.4%) A spring is in the air 4 ( 2.7%) SpyHunter4K 4 ( 2.7%) Lode Runner 4K 4 ( 2.7%) Alpha Shot 4k 3 ( 2.1%) Spline4k 3 ( 2.1%) Sail 4K 3 ( 2.1%) Lady Bug 4K 3 ( 2.1%) Gravitator4K 3 ( 2.1%) C&C 4K 3 ( 2.1%) Balls4K 3 ( 2.1%) Bad Sector 3 ( 2.1%) Asteroids 4K 3 ( 2.1%) Ares 2 ( 1.4%) Radar III 2 ( 1.4%) Minesweeper Deluxe 4K 2 ( 1.4%) Freefall 4k 2 ( 1.4%) Fish4k 2 ( 1.4%) Doom Buggy 4K 1 ( 0.7%) Warpstar4K 1 ( 0.7%) Sudoku4k 1 ( 0.7%) Star Fighter 4K 1 ( 0.7%) Rally X 4K 1 ( 0.7%) Puzzling Dominoes 4K 1 ( 0.7%) Jyruss 1 ( 0.7%) Java Rally Racer 4k 1 ( 0.7%) JSquares 1 ( 0.7%) Gumblocks4k 1 ( 0.7%) Frag 4k 1 ( 0.7%) Bungie Bill 1 ( 0.7%) AirPong4k 1 ( 0.7%) 4kanoid ( Fork-a-Noid) 1 ( 0.7%) 4k Kung Fu 0 ( 0.0%) W4K 0 ( 0.0%) Sokoban4k 0 ( 0.0%) Skeet Shoot 0 ( 0.0%) Road Kill 0 ( 0.0%) Ripples 0 ( 0.0%) R4k 0 ( 0.0%) MiniSudoku 0 ( 0.0%) LunaBug4K 0 ( 0.0%) Kana Invaders 4k 0 ( 0.0%) Kaboom4K 0 ( 0.0%) Gandalf 4K 0 ( 0.0%) Daleks4K 0 ( 0.0%) Calc 4K 0 ( 0.0%) B4k 0 ( 0.0%) 4kTetrisAttack 0 ( 0.0%) 3DTunnel

Ye, guess it werent enough votes. But at least the top 5 (or top 11) seems to be pretty accurate (albeit not distinctive enough). It also shows that the ramping is very extreme or even harsh. About as extreme as the whole make or break stuff of commercial games (eg only 5% of the PSX games made a profit).

For more meaningful results we would need at least 4 times as many votes, I believe.

#7

For meaningful results I think you want to be seeing 10x otherwise you’re not factoring in the developer community bias.

Kev

#8

BOOM! Headshot!

[edit:]
That was meant as a joke, btw. :wink: I’m not really that much of a bad winner.