Yesterday morning, in about an hour and a half starting about 10 minutes after I woke up, I rewrote the work distribution scheme for my multithreaded embarrassingly parallel election simulation program. It was ok and working before, but it's better now.

Expanding on the visualization of regions of a 2D political space where various choices are elected, I did a couple series of small multiples:
three choices, two held stationary, one moving;
four choices, varying population distribution grouping tightness.
Probably the most interesting pages of images are for Instant Runoff, here and here.

Being home sick from work today, I turned on the TV news. I don’t usually, and special for tonight I got a big dose full of New Hampshire primary coverage and speculation. I’ve been following this and the Iowa caucus by my favorite internet sources, but it’s different seeing it on the screen that presumably millions of other people are seeing too. I guess this is the reality most people are seeing.

It sucks, let’s change it.

I don’t want a 2nd, 3rd, 4th or 5th class vote. I want a vote just like everyone else, but I don’t get one. People Iowa, New Hampshire, Nevada and South Carolina get better votes than I do, and I’m bitter. It’s not their fault, except perhaps through complicit silence. If I had the same opportunity I’d sure take it.

But I’m still sick of having my vote compromised. I want to vote for Dennis Kucinich, who is in my opinion the best kind of person to have as our next President. But I was just thinking that depending on how things are going after IA, NH, SC and NV, I might compromise and vote for the best of the rest, John Edwards. That sucks, I hate it. It’s a stupid system. There are better ways. Grr!

In 2012 can we have a one-day national primary? Please? Also, let's collapse this insanely long campaign season and have it in June or July.

Imagine political opinion space is a plane. Ok, lots of people measure like liberal-conservative on social and fiscal scales, but really any axes will do. Now position some candidates on that plane. Now simulate an election with the the center of a gaussian distribution of voters at the center of pixels on that plane voting for the closest candidates. This is a great way to visualize the behavior of election methods.
( one sample image of IRV being weird )
lots of pictures here
credit due to the original here, but my pictures are bigger. :-)