Ken, thanks very much for info about instant runoff voting in SF. Easy of use is key. More coalition-building and issues-based campaigning, and less mud-slinging also sound appealing…

Robert, great points. Do you happen to have a recollection of where Mayor Perez stood at the time, on the matter of district representation, in 2002? Has he gone through a change in outlook?

Thanks for the update on the meeting. I’m very concerned about the talk of ward or district representation. As the city council seems to be shaping up a bit in their opposition to the mayor, this attempt seems to be a way to split the unity unity within the Democratic and Working families party. Furthermore, talk about district representation seems to be diverting our energy from the increasing need to question the power of the executive administration and the pressing issues at hand. The current executive was very influential in the 2002 Charter Revision, so why is the he only now interested in pushing this change? A move towards district representation would be a move from a “strong” mayor form of government to a “stronger” mayor form. Unfortunately, the district representation will be advertised as a more “democratic” system. Is it more demoractic to have a strong mayor and weaker, more dispersed city council?

Great reporting, and thanks for keeping us updated on this.

Understanding that representation is at the top of the agenda for the charter revision commission, may I suggest that you look into public financing mechanisms for the mayoral race in 2011, and instant runoff voting. San Francisco has experienced significant success utilizing IRV.

http://www.commondreams.org/views05/0112-23.htm

Thanks for your efforts.

Peace,
KK

The conclusion is that the optimal number of at-large representatives would be the square root of the total number of representatives; the rest would be ward or district representatives.

The analysis was based on optimizing an average voter’s “influence” over the workings of the city by influencing who is elected. I won’t pretend to understand the mathematical details.

Springfield’s upcoming ward setup is close — there will be 13 representatives, with 5 at-large and 8 ward. Optimally there should be just 4 at-large representatives, but it’s a great step forward.

It seems as though the plan suggested for Hartford isn’t quite so close. With 15 suggested members, the ideal at-large number should be 4, not 9, meaning there should be 9 district reps.

One thing I have noticed for a long time in Springfield is that an entire body that is at-large makes it very easy for poor performers to “hide” among the crowd. In Springfield, each race contained 18 choices, and each voter was to choose 9. This made it very hard to “throw someone out” because even if most people agreed that someone was bad, they couldn’t agree on an alternative. It also amplified the effect of “name recognition voters” because those votes tended to pool with the incumbents. (I know that’s how I voted the first few times I was eligible to vote, so I assume others do as well.)

Picture this: 10,000 voters, each pick 9 of 18 candidates. Assume 10% are name-recognition voters (which seems low), so each incumbent gets 1,000 votes and there are 9,000 issue voters.

Let’s say that 75% of the issue voters (6750) agree that incumbent A is lousy, so they don’t vote for her. 25% (2250) like her. That candidate gets 2,250 votes, plus the 1,000 low-information voter votes, totaling 3,250 votes.

The remaining voters can’t agree on who to pick in her place. They divide their votes evenly among the 9 challengers. They cast their other 8 votes for incumbents.

Each other incumbent would get 9,000 votes from the issue voters, plus 1,000 votes from the low-info voters. Each challenger would get 843 votes from the info voters, and no other votes.

The end result would be the 8 popular incumbents would have 10,000 votes, the widely-reviled incumbent would have 3,250 votes, and the challengers would have 843 votes.

And people would be left scratching their heads — because even though 67.5% of the people wanted someone voted out, she was returned to office with nearly 4 times the votes of the nearest challenger.

This explains why Springfield has seen precisely 4 incumbents turned out of office since 1988. Well, that and the fact that having to pick 9 choices makes most people into low-information, name recognition voters. That is too many people to know deep information about.

I’d like to do some more analysis of the options and the suggestions that are put forward once this initial public hearing phase is done. Still more people to hear from, hopefully, but it remains yet to be seen whether there are going to be any significantly new ideas or proposals on this subject.