Winners of the competition

[quintin3265]

I'm pleased to announce the winners of Composition Combat. Without further ado, here are the winners:

6.6412 Etiolating Shade
4.7826 Decimation
4.5972 Certain Uncertainty v1.0
4.3077 Carystas
3.9615 Absit Omen
3.7667 Dread March
3.3125 Post-Apocalyptic Game Theme
2.9545 V.A.L.K.Y.R.S
2.0417 NuN's Chaostry - New Hope
1.8421 Android

Congratulations to KiddCabbage, who won the competition. Second place goes to Muuurgh, who earns a free game. dabeat11 was the only person to submit more than one revision of a song into the competition, so she wins $30.

There are several files available online for those who would like to analyze the statistics:

-http://www.shoemakervillage.org/results/cc - average ratings.csv: the results listed above
-http://www.shoemakervillage.org/results/cc - average ratings without 1s and 7s.csv - the ratings when the 1s and 7s are dropped from the results
-http://www.shoemakervillage.org/results/cc - ratings count.csv: the number of each type of rating for each song (these results are aggregated, so that IP addresses and other information isn't revealed)
-http://www.shoemakervillage.org/results/cc - version count.csv: the number of versions of each song that was submitted to the competition.

As you can see and as was expected, there are many 7s in the results. In particular, KiddCabbage's votes consisted of an abnormally large number of 7s. However, KiddCabbage wins the competition no matter which way one looks at the results - whether all the votes are counted, whether 1s and 7s are omitted, whether votes in the same subnet are discarded, and so on. I think that the results being the same any way we look at them shows clearly that KiddCabbage is the winner.

For second place, you'll notice that "Decimation" received many more 7s than 1s. If the 7s and 1s are removed from the results, "Certain Uncertainty" would place second. Judging by the comments on the site, in these forums, and in the voting results themselves, "Certain Uncertainty" simply seems to be a song that people either love or hate. If you look at its results, it has almost a perfect inverse normal curve to its tallies. I don't see how it's possible to justify removing all the 1s and 7s from "Certain Uncertainty's" totals since people obviously felt strongly about the song one way or the other. As a result, we decided to let the raw vote tallies stand without any changes.

By the way, "Carystas" would have placed in the prizes had the voting period been even a few hours longer. The competition was literally decided by minutes.

Finally, the duplicate voting algorithm did not find any systemic cheating that invalidated the results. It may be that some competitors asked a lot of friends to vote the maximum rating, but that was not prohibited in the rules, and all competitors were able to encourage people to vote equally.

There is already talk of another competition in the works. The next time, we hope to have the "compos" feature implemented so that a lot of these calculations can be performed automatically. First, though, I want to fix bugs that I've noticed in the site, such as some of the profile image thumbnails being reduced to white squares.

Thanks to everyone who participated, and we hope to see you around the site or these forums! I'll contact the winners soon to award the prizes.