changing democracy

[bolson]

Last Friday I had a nice chat with a staffer at my state assembly member (Pedro Nava)'s office. Without an appointment, I walked in and handed her a nice flowery cover letter, 3 pages of rough draft bill text and my analysis of the June 6 Primary in which I describe how 26 races in California went to a 'winner' with less than 50% of the vote and 9 of those had a better than 40% chance that the 2nd place candidate should have won. We then proceeded to talk about the bill and how it works. She tried to play devils advocate and poke holes in what I had. We probably talked about this stuff for ten or twenty minutes. I don't know how far it will get with just that. I'll probably have to call back every couple weeks for the next year to try and poke it along.

I also mailed a copy off to my state senator, Tom McClintock, since I didn't feel like driving an hour down to Thousand Oaks. I'm curious to see what if any response I get from this.

Comments

single transferrable vote

[cos]

Interestingly, your definition of Single Transferrable Vote is significantly different from the way STV works here in Cambridge, where we use it to elect the city council and school committee. The mayor is elected by the council from among them, so STV is how we elect all of our city's elected officials.

Some analysis of Cambridge elections here.

Re: single transferrable vote

[lightcastle.livejournal.com]

It's unclear to me how STV works in Cambridge, that page doesn't actually seem to say. So how is the Cambridge version different?

Re: single transferrable vote

[soong.livejournal.com]

Based on this page it looks like the difference is mostly "random transfer" vs "fractional transfer". I specified a fractional transfer method which works best with a computer count, but the STV method they've been using "for over 60 years" can be done reasonably by hand.

Re: single transferrable vote

[cos]

There's also a difference in how you describe the rounds. In Cambridge, after the initial round where any candidate who got over quota in #1 votes is elected, and the second round, where all candidates with fewer than 50 are eliminated, each round is the elimination of exactly one candidate. Their transfer votes can elect 0, 1, or more other candidates, but that doesn't affect the fact that the next round will again be the elimination of exactly one candidate. The process continues until all seats are filled.

The "random transfer" in Cambridge is actually not random, nor is it stack or queue. It's a systematic transfer of whole votes rather than a wholesale transfer of partial votes. The randomness is in the order of ballots cast: to transfer 1/3 of the surplus ballots, for example, transfer ballot #1, #4, #7, etc.

Re: single transferrable vote

[soong.livejournal.com]

Fractional transfer is better than (psuedo/semi/effectively) random whole transfer because it's more fair. In whole transfer, it's up to the luck of the draw whether your vote was 'wasted' on voting for the guy who was too popular and won by too much or whether your vote transfers to get you something else of what you want. Whole transfer is easier to do by hand, but it's not hard to find an open source STV implementation.

The exact definition of the rounds can also sometimes affect the outcome, but I think STV always follows the basic pattern of: elect if you can, disqualify until you can elect some more.

Re: single transferrable vote

[lightcastle.livejournal.com]

Everywhere I know that has recently tried to get STV installed has gone the fractional transfer route.