Except that we don't select the President by county, so the second map is kind of irrelevant. You have huge areas in the West and Mid-West where few people live. The map distorts this and makes it look like the 7 million people in NYC are the same as the 700 people in Little Crossroads Montana. The fact remains that more than 50 million people voted for Kerry, their geographical distribution and density is not relevant. We don't elect a President by land area.

Sounds like you're assuming I am trying to make some kind of partisan statement. That's not the case. The map doesn't distort anything. Nowhere does the map or I say that this is how we elect people. My point was that it's interesting to compare the two. Had I posted just the county-by-county map and pointed to it as being relevant to how we elect people, *that* would be a distortion. But that's not what I did.

And, in a way, we do elect by land area. Sort of. At least in part. It has relevance in the electoral college. Hence the state-by-state map.

What the county map shows is that the democratic party majorities are for the most part centered in metropolitan and a few other key areas, and that the vast majority of rural America swings republican. No surprises there. We all know that if we instead used blue and red dots for each person, we'd see varying shades of purple across the map.

Come to think of it, that might be interesting to see. Any maps like that?

Ah HAH! Ask for varying shades of purple and what do you know...

http://www.cscs.umich.edu/~crshalizi/election/

- gh