November 09, 2004
Electoral Power Continued
I've written before about a study I completed that shows how much electoral power each candidate had through the final sixty days of the campaign and election day.
(Click to enlarge.) Election day ended up with Bush having about 275 EVs worth of electoral power, and Kerry having about 258. The spike at the end is all the undecideds breaking for their candidate, compared to the polls.
The way I figure this is by giving each candidate a portion of every state's electoral votes, by how much support that candidate had. This makes every state award its EVs proportionally, and is a more accurate gauge of the actual support each candidate has - it represents the underlying physics of the Electoral College.
One of the most common points people have made against this technique is that it exaggerates the power of rural voters in safe Bush states. The critics say that since these small states have so little population, then the Bush voters in this state are given too much weight, so this method exaggerates Bush's level of support.
It is true that the Electoral College gives rural states more voting power than the most populous states. But in order for that critique to have merit, it would have to mean that this study gives more of an advantage to these Bush-state voters than the Electoral College does itself.
The best way to gauge this would be by looking at the safe states. If this were true, my method would exaggerate the Bush support in the safe states.
So, I took a look.
We can consider the safe states to be the states where the margin of victory was greater than ten percentage points. Here's how those states added up in the Electoral College:
Bush: 183
Kerry: 146
Now, let's look at how those states would have played out had we awarded the EVs proportionally:
Bush: 171.3
Kerry: 154.4
So, that should be enough to smack that criticism down. Let's look at it again if we look at the states was greater than 5%. First, the Electoral College:
Bush: 249
Kerry: 183
And, the proportional study:
Bush: 223.2
Kerry: 204.5
The reality is that the Electoral College exaggerates Bush's support even compared to this proportional study. It's actually pretty obvious once you think of it - since there are so many small states in the Electoral College, of course the winner-take-all nature is going to work in Bush's favor. In this election, Bush actually outperformed the base numbers due to the inefficiencies of the Electoral College, much like how Gore outperformed them in 2000. My graph, above, actually understates the gap between the two candidates. The truth is that Bush actually had more electoral support than Kerry all along, and we have a ton of work to do before we can have an even chance of beating these folks consistently.