Monday, November 08, 2004

Electoral Math

So I hear from a couple of folks a renewed call to get rid of or alter the electoral college. The thought around here is that it disproportionately helps the large, but sparsely populated, states and therefore the Republicans. It does this by giving each state a vote for each of their two senators. So a state like Montana which only has enough people for one representative gets 3 votes. The questions of interest are 1. What is the magnitude of the effect and 2. Does this effect commonly help Republicans? Click "read more".

Let's start with 2000. In 2000, Bush carried 30 states; Gore carried 20. The final EV count was 271 to 266 (with one abstention that would have gone to Gore had it mattered.) I am giving that one to Gore (267).

In the popular vote Gore lead 50.99 million to 50.45 million. So Gore's share of the 2 way popular vote was 50.3% while Bush's was 49.7%. But Bush's share of the electoral vote was 50.4 to Gore's 49.6. In an extremely close race, the electoral college can produce an odd result.

What if we eliminate the senatorial votes from the electoral college? Now Bush gets 60 fewer EC votes and Gore gets 40 fewer. New total: Gore 227, Bush 211. This gives Gore 51.8% of the EC vote and Bush 49.3.

In 2000 the electoral college and in particular the "extra" senatorial representation gave Bush the victory.

Also, we now have a template:

2000 EVs EV% Pop% EVsAdjusted EVAdjusted% Difference (EV and Adjusted)
Bush 271 50.4 49.7 211 49.3 -1.1
Gore 267 49.6 50.3 227 50.8 +1.1 (rounding)

Let's apply it:

Bush 286 53.2 51.5 224 51.1 -2.1
Kerry 252 46.8 48.5 214 48.9 +2.1

So Bush was helped but would have won regardless. What about some other races of note?

1992 EV EV% Pop% EVAdjusted EVAdjusted% Diff
Bush 168 31.2 37.7 132 43.1% +11.9
Clinton 370 68.8 43.3 306 56.9% -11.9

Two real lessons here. One: EVs usually help the winner, not the Republican. Two: the Electoral college really hurts 3rd party tries, not the dems. Remember, Perot picked up 19% of the vote and got no EVs.

Interestingly, in 1976 both Carter and Ford got 25 states which produces the statistically unlikely effect of having the EC help the loser slightly.

One final one for you Gore 2000 fans:

Hughes 254 47.8 48.3 218 50.1 +2.3
Wilson 277 52.2 51.7 217 49.9 -2.3

So, absent the "senatorial advantage" of the smaller states, Wilson would have lost in 1916.

Overall, it doesn't seem like the "senatorial advantage" is large enough or consistant enough to present a problem. It has had a seen effect on two elections (Hayes-Tilden 1876 and Bush-Gore 2000) and an unseen effect on a third (Wilson-Hughes 1916). One out of every 18 elections we have held since 1788 and only two in the modern era with one breaking for each party. Not a bad record.

This has to be weighed in the balance with the positive effects of an electoral college system (to be discussed in a future post.)


Blogger sweet-ness said...

Captivating blog. I love surfing the web for the
type of blogs that you do. It had me on the edge of my
seat and I kept going back to again and again!
Jump into my pay day loan cash advance blog.

4:11 PM  

Post a Comment

<< Home

Listed on Blogwise