One Man's Electoral Analysis

With the election less than two weeks to go, I have increased my immersion in Internet campaign reporting and opinion to saturation levels. And like a soaked sponge held above this page, I'll be dripping my thoughts all over the blog in the next couple weeks.

Lately, I have been very curious about how the electoral college is going to play out. The media is obsessive about national polls, but national polls aren't totally relevant to the final outcome--arguably not relevant at all. The media also periodically offers a random assortment of polls from "battleground" states, without any context to what those polls mean to the larger electoral picture. What does it mean that a new poll has Obama up three in Ohio, and McCain is up five in Georgia?

In order to ease my confusion, I spent a couple hours researching the latest state polls and various scenarios and trend lines and whatnot. I now think I have a clearer picture of how things will play out, though of course, we are always one heart attack or racial slur away from a total upending of everything. Bear in mind that I am bad at math and don't have a terrific handle on state abbreviations, so there may be some errors. Overall, though, I think the portrait is accurate.

First things first: 538 total Electoral Votes (EVs), 270 needed to win.

Obama has 213 EVs essentially banked, McCain has 142. Certain states just aren't in play and don't seem to have any prospect of being in play. Given recent trends, it seems Obama *might* be able to pull from McCain's banked EVs, the opposite scenario less likely, but given that the election is thirteen days away, it's not probable that either candidate will overcome a 10+ polling advantage in places like Texas or New York.

In addition to each candidate's banked EVs, it seems *very likely* that Obama will win Washington, Minnesota, New Mexico, and Pennsylvania, for an additional total of 47 EVs, bringing him to a highly probable total of 260 EVs.

McCain seems to have Georgia and North Dakota locked up, giving him another 18 EVs, for a probable total of 16o EVs.

In other words, both candidates can rely on 260 (Obama) and 160 (McCain) EVs going in to election night. That means of the remaining states, Obama needs to find 10 EVs, McCain 110. So at this point, Obama has a real advantage.

Here are the non-banked, non-very likely states, along with their current aggregate polling numbers and EVs.

STATE / EVs / POLL LEADER
NH / 4 / O +5.8
IN / 11 / M +3.5
MT / 3 / M +3.7
OH / 20 / O +.7
MO / 11 / O +1.6
NV / 5 / O +2.4
FL / 27 / O +2.9
NC / 15 / O +3.1
CO / 9 / O +5.7
VA / 13 / O +8.5

Based on these polling numbers, if the election were held today and each state fell as these polls suggest Obama would pick up an additional 115 EVs, and McCain 14 EVs. Their grand totals would be 364 Obama and 174 McCain--an Obama blowout.

So, just to avoid a blowout, McCain needs to make up ground. Now, granted, the above states (many of them, at least) have narrow enough margins that you'd think McCain could pick some off. But, again, he needs 110 EVs to win, and there are only 118 EVs available in that group. So McCain needs to basically run the table to win.

To put it another way, McCain could win Florida, Ohio, New Hampshire, North Carolina, and Virginia and still lose. He could win those states, and hold the states where he's leading, and lose.

Another perspective--if Obama holds his banked and highly probable states, he needs one state of 10+ EVs to win. If he holds Virginia (where he is way ahead), he wins. If he picks off Florida, he wins. Missouri, he wins. Ohio, he wins. Again, these are all states where current polls show Obama ahead, and if he wins one of these and loses every other swing state, he wins.

This is why pundits are now saying McCain's path to victory is very narrow--it is really, really narrow. He needs to win every 10+ EV battleground state, including ones where he is currently trailing, he needs to hold Indiana, and then scrape together another 13 EVs from the remaining states.

Let's circle back to the beginning at this point. I stated that Obama had in his "very likely" column Pennsylvania. Despite the fact that McCain trails very badly there (down 15.3 in aggregate polls--hence the Very Likely Obama status), McCain & the GOP are recently committing a lot of time and money to the state, while pulling out of theoretically swing-able states like Colorado. This makes sense given the broader electoral picture. If McCain can shake loose these 21 EVs from Obama's pile, it gives him a ton more flexibility in the rest of the map. He can lose some of the bigger states, and still win. He probably thinks, at this point, that he has a better shot claiming Pennsylvania than running the table in the closer states.

FINAL ANALYSIS: The keys to this election are Pennsylvania and Virginia. If Obama wins these two, he is virtually (mark the word, as Dan Rather would say, virtually) assured of victory. He can win without the help of wild cards like Florida and Ohio, which, given their history, seems important. If McCain takes Pennsylvania, he is very much back in the game electorally. If he doesn't, he needs to pull out Virginia, along with every other big swing state (and more or less all of the smaller ones, too) to get to 270.