Ever since the 2000 fiasco, I’ve been thinking about the electoral college. I had pretty much settled on the idea that the best route would be not to try to abolish the electoral college itself, simply because of the challenge of passing a constitutional amendment, particularly over a kind of esoteric issue. Rather, I have thought that people should work within their states to shift to a more proportional allocation of electoral votes, assuming that passing a state initiative would be a lot easier than amending the constitution.

Well, as you may or may not know, Colorado put a measure of this sort on the ballot this year, and it got clobbered pretty soundly. I think that positioning it so that it would affect the electoral vote the same year of its passage was a bad idea, and clearly from reading coverage like this Denver Post Editorial, many people still believe that they have a lot to lose by changing the system unless all the states change at the same time.

Anyway, I have been wondering about how some of this might pan out, and I found a spreadsheet of results at electoral-vote.com, so I fiddled with it a little to see some numbers. I set up two scenarios. One is like what I believe Colorado proposed: divide the electors directly according to the proportion of the popular vote won by the candidate. In this scenario, I came up with the Electoral College going to Bush, 285-250. (Three votes were lost in rounding somewhere.) Because I don’t have data down to the congressional district level, I couldn’t directly figure out how it would work in the Maine/Nebraska scenario. Those states allocate one elector per congressional district to the winner of the popular vote in that district, and then award the two “senatorial” electors to whichever candidate has the overall majority. Given that neither Maine nor Nebraska has actually ever split their votes, my methodology is probably flawed, but since I could get the spreadsheet to do the hard work, I set it up to divide all but two of the EC votes proportionally, and then add two to whichever had the overall majority vote. Now that I think about it, this is totally unlikely to be meaningful. Because in most states, the districts wouldn’t split proportionally — again, it’s never happened in Maine or Nebraska.

The long and short of it is, this time, Bush won the popular vote, so he would have won in most more direct methods of election too. It would be wrong for me to only hate the Electoral College when my candidates lose, but to think it’s the right way if it favors my candidate.

In case anyone feels like fooling around with the numbers, here are data files:

If anyone knows where I can find 2000 numbers by congressional district, maybe I’ll see if I can run them more accurately.