|So far in 2008, over 400 polls that measure national voter preference for the inevitable Obama vs. McCain general election have been released to the public. According to Pollster.com, during this time the two candidates have never been separated by more than 4%, indicating a third consecutive close election. Despite the overwhelming Democratic advantage in other electoral areas, this is not very surprising. After all, John McCain has long held an extremely high favorable rating, and is currently the most favorably viewed Republican with a national profile. By way of contrast, Barack Obama is a relative newcomer who is seeking to become the first African-American President of the United States. In all likelihood, it is only in a year that extremely favorable to Democrats, such as 2008, that Barack Obama would have any chance at all against John McCain.
While the national polling picture points to an extremely close election, national results do not determine the winner of the Electoral College. Instead, in an election this close, the eventual winner will be the candidate who can win the most electoral votes in the closest swing states. Unfortunately, while there is more than enough national polling data to produce an accurate measurement of the national popular vote, very few public polls have been released among swing states. In fact, over the past thirty days, only five states, California, North Carolina, Ohio, Pennsylvania and Texas, have been blessed with public polls from three or more polling organizations. As such, our ability to measure who is currently ahead in the Electoral College is limited by a lack of data.
With few state level polls, in order to determine the coming shape of the general election, we are forced to turn to other data points. Taking the partisan voting index from the 2004 general election, and combining it with the demographic regression on fivethirtyeight.com, the following states are projected to be the six decisive, uber swing states for the 2008 general election:
- Colorado (9 electoral votes, projected +0.1%): 2004 PVI: -2.2%, 538 regression +2.0%.
- New Mexico (5 electoral votes, projected+1.5%): 2004 PVI +2.7%, 538 regression +0.2%
- Ohio (20 electoral votes, projected +1.6%): 2004 PVI +0.3%, 538 regression +2.9%
- Nevada (5 electoral votes, projected +2.3%): 2004 PVI -0.1%, 538 regression +4.6%
- Wisconsin (10 electoral votes, projected +2.5%): 2004 PVI +2.9%, 538 regression +2.1%
- New Hampshire (4 electoral votes, +2.6%): 2004 PVI +3.8%, 538 regression +1.3%
The next closest states, in order, are #7 Florida (-3.1%), #8 Iowa (+4.0%), #9 Pennsylvania (+4.2%), #10 Virginia (-4.7%), and #11 Michigan (+4.9%). However, the unusual organizing situations in these states scramble the matter somewhat. By not campaigning as extensively in Michigan and Florida during the primary season, Obama should have a more difficult time winning in those two states than in the others listed here. By the same token, McCain should struggle, comparatively speaking, in Iowa and Pennsylvania, due to his relative lack of campaigning in those two states. These estimations are born out by current polling, which currently shows Obama stronger than expected in both Iowa and Pennsylvania, but McCain stronger than expected in both Florida and Michigan. As such, Michigan and Virginia move past Florida, Iowa and Pennsylvania in the rankings of decisive swing states, to become #7 and #8 respectively.
While there are other states that are close, realistically, only the top eight, and possibly only e top six, will be decisive. For example, while Obama leads by single digits in Iowa, Pennsylvania, and Massachusetts, if he were to lose any of those states it is extremely unlikely that he would be able to win enough of the eight closest states to reach 270 electoral votes. By the same token, while McCain only leads by single digits in Florida, Indiana, Missouri and North Carolina, it is extremely unlikely, if not impossible, for him to win enough of the eight closest states to reach 270 electoral votes. These more outlying swing states are all actually integral to winning coalitions for either Barack Obama or John McCain. As such, if we ever get to a point where Obama is leading in one or more of the four McCain states listed in this paragraph, or McCain is winning in one of the three Obama states listed in this paragraph, then the outcome of the election will be obvious and we won't need to examine electoral math minutia in order to project the leader / winner.
Looking only at the eight states that are currently projected to be decisive, we quickly arrive at a much clearer and easy to understand picture of the general election:
Obama Integral States: 228 Electoral Votes
Currently Lean Obama (48): Colorado (9), New Hampshire (4), New Mexico (5), Ohio (20), Wisconsin (10)
Currently Lean McCain (35): Michigan (17), Nevada (5), Virginia (13)
McCain Integral States: 227 Electoral Votes
Overall, Obama currently holds a narrow 276-262 edge in the Electoral College. However, the lead is far from solid, and is well within the margin of error.
Now, I'm not arguing that the other forty-two states don't matter, and that Democrats and progressives should only focus on Colorado, Michigan, Nevada, New Hampshire, New Mexico, Ohio, Virginia and Wisconsin this year. Simply as a matter of strategy, even if your objective is only focused on eight states, it does not make any sense to place all of your resources in only those eight states. Not only would such a narrow focus create a real image problem for the Obama campaign, but it would make it easier for an under-resourced opponent like McCain to pool his resources and stay competitive. Further, the point is not just to win this election, but to win in a way that builds a governing majority once in office. In order to achieve that end, it will be necessary to have a thriving Democratic Party in all, or nearly all, 50 states, and deliver vast downticket wins. If you want to govern the entire United States, you need to campaign in the entire United States.
What I am arguing is that if you want to know who is winning the general election, or if you have the opportunity to spend either personal or organizational resources on the general election, right now you should focus on Colorado, Michigan, Nevada, New Hampshire, New Mexico, Ohio, Virginia and Wisconsin. That list of states might change as more polling data becomes available later in the campaign, but right now those eight look to be decisive.