Majority Acceptable Score voting: Difference between revisions

imported>Homunq
No edit summary
 
(4 intermediate revisions by 2 users not shown)
Line 1:
Majority Acceptable Score voting works as described below. Technically speaking, it's the [[graded Bucklin]] method which uses [[Three-level ballots|3 grade levels]] and breaks median ties using [[Score voting]].
 
# Voters can support, accept, or reject each candidate. Blanks count as 2/3 of a rejection and 1/3 of an acceptance (so 75% blanks counts as 50% rejections).
Line 7:
# Highest points wins.
 
Step 2b probably doesn't matter, because any majority-supported candidate that exists would almost certainly win in step 4 anyway. But step 2b is part of Bucklin voting, which was used in over a dozen US cities during the Progressive era. Also, it lets you say the whole method in one sentence, ifusing the personidea you're talking to understandsof medians: "choose the highest score among the candidates with the highest median".
 
Here's a google spreadsheet to calculate results: [https://docs.google.com/spreadsheets/d/1siFG6XmOZokygY-86EhAKgv8YwzKtTET6AJopyXRqu0/edit#gid=0]. On page 1, it has some examples of how different combinations of ratings would come out, suggesting that it could work well in both [[chicken dilemma]] and [[center squeeze]] scenarios. On page 2, it has some hypothetical results for the Egypt 2012 election, showing that this system could have elected a reformer over Morsi, despite vote-splitting among the various reformers. IRV could have elected Morsi. (Note: the spreadsheet does not actually check step 2b.)
Line 49:
!bgcolor="#fff"|Nashville
|bgcolor="#fff"|26
|bgcolor="#fff"|37
|bgcolor="#fff"|0
|bgcolor="#fff"|370
|bgcolor="#cfcfff"|27.474
|bgcolor="#bfbcfc"|9849.67
|bgcolor="#bfb"|76.3
|-
!bgcolor="#fff"|Chattanooga
|bgcolor="#fff"|15
|bgcolor="#fff"|3017
|bgcolor="#fff"|21
|bgcolor="#fff"|42
|bgcolor="#cfcfff"|49.926
|bgcolor="#ffffcc"|6559.17
|bgcolor="#fff"|37(75.7)
|-
!bgcolor="#fff"|Knoxville
|bgcolor="#fff"|17
|bgcolor="#fff"|2815
|bgcolor="#fff"|42
|bgcolor="#fff"|1326
|bgcolor="#fcc"|5259.87
|bgcolor="#fff"|(6477.27)
|}
</div>
 
Memphis is explicitly given 0 by a majority, and is eliminated. Chattanooga and Knoxville are both given 0 by a majority implicitly, so they are eliminated. Of the remaining two, Nashville has a higher scoreremains and wins.
 
If Memphis voters tried to strategize by rating Nashville and Chattanooga at 0 in the above scenario, it would take a bit over half of them to successfully execute the strategy. Even if all the Memphis voters strategized, Chattanooga and Knoxville voters could protect Nashville against this strategy asif long as under half2/3 of thosethem who had givengave Nashville a blank above switched to giving it a 1 (or a 2). Note that the offensive strategy involves moving a natural 1 down to the extreme value of 0, but the defensive strategy only means changing a lazy blank to a natural 1 (not to the extreme value of 2).
 
[[Category:Graded Bucklin systemsmethods]]