View Single Post
Old 02-19-2013   #1882
MascaraSnake
Retired
 
MascaraSnake's Avatar
Default

Quote:
Originally Posted by Ice View Post
Though, I do have to say, I'll miss the top-ten leaderboard thing. Will we have one of those? Or do they just not work with this new system?
Well, purely theoretically, we could divide each entry's overall score by the number of voters and n-1 (n being the number of maps in the division). That gives us a number between 0 and 1 that you could interpret as a percentage and compare between maps from different contests. Of course, this would be completely useless in practice because the number tells us nothing about the absolute quality of a map; it depends heavily on the number and quality of the other entries in the division.

Let's take two previous contests as an example: The first is March/April 2011, where the single player division consisted of two terrible maps. The second is January/February 2012, where the single player division consisted of two great maps. Now imagine that these contests had been held with the new voting system. It's obvious that the "averages" as outlined above of both maps have to add up to 100%. That means that the closer the two maps are in quality, the lower the "average" of the better map will be. In March/April 2011, the gap between the two entries was 0.25 points. Because the range of scores was pretty narrow, the results probably would have been similar in the new voting system. For example, the better map could have gotten 63% and the worse one 37%. In January/February 2012, the gap was only 0.08. The votes were more divergent, so the new voting system might have given a different result, but let's say it would have been similarly close. In that case, we could except the "averages" to be closer, like 53% to 47% or something. If we construct a leaderboard from this, the (terrible) winner from 2011 is in first place while the (great) winner from 2012 is in second. This is obviously nonsensical.
MascaraSnake is offline