by carbocation, Thu Mar 13, 2008 at 02:15:56 PM EDT
So far this year Obama's leaked spreadsheet has had a reasonable amount of accuracy in projecting the vote percentages of future contests. It's likely that Obama will end up with a pledged delegate lead. What about the popular vote? In this projection I have used the leaked spreadsheet percentages to predict the final popular vote.
80% of the Kerry popular vote from the 2004 general election was used to determine turnout in each upcoming state. In the case of Puerto Rico I used 80% of the popular vote in the 2004 gubernatorial race. For Guam I multiplied the 2008 Virgin Islands turnout by 150%.
Why 80%? It's an arbitrary choice between two relative extremes. If you look at Ohio, its turnout was exactly 80% of the Kerry 2004 vote. Texas is on the high end, where turnout was almost 100% of the Kerry vote. States like California and Illinois had a turnout of roughly 69% with respect to the Kerry vote.
In my opinion Michigan and Florida are going to get revotes. Recent polls show Obama and Clinton head to head in Michigan, and Clinton leading substantially in Florida. For the sake of argument I've given Clinton a moderate 4 point lead in Michigan, and a substantial 17 point lead in Florida. This seems like a best case scenario for Clinton, as she her largest "big state" blowout was also 17 points in New York.
In this scenario Obama would end up with a 95,573 vote lead. Puerto Rico is a large an unknown. If it the states breaks down PPD and PNP lines, then I doubt either candidate will net a substantial popular vote. Florida and Michigan are the other wild cards. If they aren't considered, then Obama would have a popular vote lead of roughly 800,000 votes. Even in a favorable Clinton scenario as I've described, she would not be able to overcome Obama's lead with wins in Michigan and Florida. I have a feeling that the popular vote from these two states is greatly overestimated with the method that I've used, as one state will be using 'firehouse primaries' and the other will be using an untested mail-in ballot system.
Here's a graph of Obama's net popular vote lead over time. I think this well illustrates the mountain Clinton will have to climb in order to win this nomination:
I can understand why Clinton has decided to stay in this race. However if she does not over perform this prediction in Pennslyvania, North Carolina, or Indiana then I believe her chances for overcoming Obama's popular vote lead are drastically reduced. Even as an Obamaniac, I don't see a 140-150 pledged delegate and a 95,000 popular vote lead as convincing enough to force Clinton to drop out before the convention. Unfortunately, all signs point to Denver.