Hunting The Muse

2000 Proportional Results

I was bored so I decided to re-examine the 2000 Presidential Vote using some numbers I found here.

The biggest statistic that people always mention is that Bush won even though Gore won the popular vote. And then all the gnashing of teeth about Florida.

I've long held the opinion that the popular vote argument is stupid. I'm actually in support of the general concept of the electoral college. We have a nation with a history of balancing popular representation with regional representation. Even our congress is set up that way - the Senate gives states equal representation, while the House is (in theory) set up for popular representation. So it makes sense to me that the electoral college would strike the same difference, of giving each state different weightings to balance its popular representation (by awarding it one electoral vote for each congressional district) with its regional representation (by also giving each state one electoral vote for each of its two senators).

So, given that the general concept is a good one, how good is our actual implementation of it? One of the common arguments against the electoral college is that each of the states are "winner take all", so for instance none of the democrats in Alabama are having their votes "counted" in the electoral college calculations. In general I believe this balances out because other states that go Democratic will then not have their Republican citizens "count".

But what would be more fair? Leaving aside the question of how we determine how many congressional districts (and electoral votes) each state receives.

One thing that would be more fair would be to proportionally award the electoral votes, by how much popular support each candidate received in each state. If the real purpose is just to better represent states that have regional interests but not as much population, then it really is just a matter of weighted averages. So if Texas had 32 electoral votes, and Gore had 38% of the Texas vote, then he'd get 12.16 electoral votes.

So I went through and calculated it for each state - and rather than figure it in terms of electoral votes, I figured it in terms of person votes. Since the smaller states have more regional weighting, the votes of the people in those states effectively count for more than they do for people in California. By weighting all the votes by how much their states are worth, I came up with some interesting results:

Of the 105,412,329 votes cast in 2000, when the votes are weighted properly for regional representation, Bush actually got more:

Bush: 48.17%

Gore: 48.01%

Nader: 2.77%*

Other: 1.05%*

This "weighted popular vote" is even closer than the regular popular vote we had, where Gore led by 0.5%.

What's interesting is that if things had been more fair, Bush would have won. But we all know Bush won unfairly. Why was Gore in the position to have been the rightful winner, even though the intent of the system would have picked Bush?

Believe it or not, it's because Gore's strategy worked better. Of the top seven closest states, Gore won five of them. The other two were New Hampshire, and of course, Florida, which Gore actually won. What this means is that the argument of "I'm an x in a y state so my votes didn't count" actually worked overwhelmingly in the Democrats favor in 2000. Basically, Gore squeezed a lot of blood out of the turnip of 2000, with the Nader distraction, an apathetic Democratic base, and a population that underestimated Bush's conservatism.

A postscript: What the Democrats should really be doing is convincing 30,000 California Democrats to move to the east side of Lake Tahoe. I mean geez, just live in your vacation home for three months next winter, long enough to register, vote, and move back. If the states are the same as 2000, you'd swing the whole election.

Posted by Curt at December 7, 2003 03:12 AM