Which of these six?

This blog entry is related to the poll:Which Election method do you like? (Total: 3 posts)
It is discussing the poll topics:
  • Instant Runnoff Voting (1 post)
Other topics:
Project

For public elections, the vote needs to be simple. IRV can be overdone and underdone. The Australians require a full ranking of candidates. With four or more good candidates, it's a long hard count. Voters for the Mayor of London are limited to 2 picks. Very countable. But in their most recent election more than 10% of the voters didn't pick the top two. They didn't count.

I favour a ranked choice ballot where voters may rank 1, 2 or 3 of the choices. It is hand countable if done efficiently. Vote counters can go home after the ballot totals are recorded, long before the runoff process starts. In matters of left to right politics, it shows which way a voter, who is willing to compromise, is leaning. Sometimes it is "in the middle, then left and then back over to the right". That is significantly different from a vote cast "in the middle, then left, and then even farther left". Votes like these die in London.

I am a strong supporter of the "later-no-harm" criteria in judging the fairness of a voting system. I don't like it when my own vote for my lesser preference helps my higher preference lose. I prefer that voters cast an honest opinion without compromise. Therefore, all these other systems cause me grief. In range ... I can't give anything more than zero to a less preferred candidate or I could be helping my first choice(s) lose. In Emocracy and approval, I must give all my exactly equal highest preferences the maximum and every other candidate the absolute minimum, or I could be helping my first choice(s) lose.

IRV protects my first choice. Unfortunately it can blindly go too far. It's called the spoiler effect. Condorcet can fix that, but it is so complicated and a nightmare to count by hand. With all these concerns, I will reluctanly support any version of IRV over the rest. However, I prefer a version of IRV with a three vote maximum and a mini Condorcet style twist which I call IRV+ (described in my personal book).

Comments

more choices

Tags:+Later-no-harm criterion

The poll is not necessarily limited to the current 6 choices. More can be added any time. If any are, it will be made obvious in your polling station.

Election method criteria will obviously be a recurring theme in this site. I've added a wiki page on "Later-no-harm criterion" which currently links to the already well developed wikipedia article. If members here have original ideas not found in wikipedia, we can add them in our wiki without worrying about wikipedia's policies regarding 'notability' or 'original research'.

We could create a poll on election method criteria, see the ones we feel are the most important.

IRV+:
http://en.minguo.info/book/ran_in_93/irv