Nate Silver: McCain's (long) road to electoral win

[w¡n9Nµ7 §£@¥€r 123]

With nine days to go until the election, John McCain's electoral map is in tatters. According to Pollster.com, Obama presently holds leads of five or more points in 23 states containing 286 electoral votes - 16 more than he needs to clinch the electoral college. Obama holds smaller leads, moreover, in another seven states containing 92 electoral votes, including places as far afield as North Dakota. Obama has even led some recent polling in Georgia, West Virginia, and Montana.

This situation is a reflection of the failures of McCain's strategy. During the summer, he poured millions of dollars into attack advertising rather than building up the sort of robust ground operation that won George W. Bush the presidency in 2000 and 2004. He underestimated the threat in Virginia, North Carolina and Indiana until it was too late. And now, he and Sarah Palin are jetting all around the country like chickens with their heads cut off - falling into exactly the trap that the Obama campaign set for them.

In order to have a chance of winning the Electoral College, McCain will need to close the popular vote gap by at least 6-7 points nationally. I am not about to advise him on how to do that, and frankly I am not sure that it can be done. Our latest estimates at FiveThirtyEight.com give McCain only about a 5% chance of pulling out a victory.

If McCain is able to close this gap somehow, however, the electoral map will look quite a bit different - and quite a bit more favorable to him. Essentially, McCain needs to subtract 6 points from Obama's margins in every state, and proceed from the assumption that this is what the map will look like on Election Day. If he is able to make that leap of faith, McCain will find it easier to pick his battles, focusing his efforts on no more than six or seven states. If I were advising the McCain campaign, I would suggest he do the following:

1. Abandon Pennsylvania, Wisconsin, Minnesota, Iowa. The notion that McCain is going to win Pennsylvania is folly. He has not led a poll in the Keystone since April, essentially since the Democratic primaries were competed and Obama laid down roots in the state. Current polls have McCain trailing by margins ranging from 7 to 13 points, well more than his disadvantage nationally. Pennsylvania is an expensive state to compete in. And contrary to the conventional wisdom, Obama did not particularly underperform his polls during the primaries (the final Pollster.com average projected Obama to lose Pennsylvania by 7.6 points, and Obama lost by 9.1, essentially within the margin of error).

Iowa is even worse for McCain; his position on ethanol is a non-starter there, and he hasn't led a poll in the state all year. Wisconsin and Minnesota aren't much better. Wisconsin borders Illinois and has same-day registration, which will allow Obama to run up the score with students in Madison and Milwaukee. Minnesota, in spite of being one of the few places where the Republicans have outadvertised the Democrats, has merely bent but not broken; McCain might get his margin within 2-3 points there, but it's hard to imagine him winning such a traditionally blue state.

2. Attack New Hampshire and New Mexico. On the other hand, New Hampshire and New Mexico might present more appealing opportunities. Obama's gains in the post-Lehman universe have come principally from white voters, which means that New Mexico, the most Hispanic state in the country, has drifted closer to the electoral tipping point. It is also dirt cheap to advertise in. New Hampshire is not as cheap, since its television market overlaps with Boston, but this is a state where McCain overperformed during the primaries in both 2000 and 2008, while Obama did just the opposite. And McCain's tax message might sell well in such a notoriously libertarian state.

3. Defend Colorado, Virginia, Nevada, Ohio and North Carolina. McCain faces uphill battles in Colorado and Virginia, where the demographic winds have shifted against him, and where he has been vastly out-organized on the ground. But they represent Obama's path of least resistance to 270 electoral votes, which means that McCain needs to do everything in his power to block it. Should Obama win all the Kerry states, plus Iowa and New Mexico, he only needs one of Colorado and Virginia to clinch the Electoral College, and right now he has solid leads in both.

McCain's problem in North Carolina and Nevada is that those states are already voting, and that Obama is banking votes there every day.

Between Clark and Washoe Counties, which represent about 85% of Nevada's population, Democratic early voters have outnumbered Republicans by about 2:1. Similar numbers apply in North Carolina, where Obama has already established a lead of several hundred thousand votes.

Ohio is more difficult to read, with polls showing everything from a 2-point McCain lead to a 14-point edge for Obama. But this is a state that is immensely dissatisfied with the Republicans establishment, and where Democrats have made huge gains in voter registrations. Unlike in 2004, moreover, Ohio has a Democrat as its Secretary of State, so something like a recount or a dispute over ballot access is more likely than not to be resolved in their favor.

4. Gamble on Florida, Missouri and Indiana. But McCain quite literally cannot afford to compete everywhere. In certain states, he needs to throw caution to the wind, and simply hope that they come back into his column if and when the national polls tighten. Florida is a good example. The state and local Republican parties are well-organized there, and Republicans have historically outperformed their polls there on election day (George W. Bush, ahead of John Kerry by only 1-2 points in most public polls in 2004, wound up winning by 5). There is also a gay marriage initiative on the ballot, which might draw older voters to the polls.

Indiana and Missouri are in some senses riskier; they border Illinois, so they'll be flooded with Democratic volunteers on Election Day, and Obama closed strongly in both states during the primaries. But Missouri is no longer a true bellwether; it was between 4 and 5 points redder than the country as a whole in the last two election cycles, and McCain has to hope that the same holds this year. And in Indiana, a state where both public and private polls have diverged wildly from one another, McCain simply has to hope that George W. Bush's 21-point advantage in 2004 is too much for any Democrat to overcome in one year.

By no means would this strategy make a victory likely for John McCain. He needs to find some way to win several news cycles during the last week of the campaign, and then he needs hope, faith, and a lot of luck. But faced with a dire situation, McCain needs to re-find the focus and discipline that he has been lacking for much of the campaign; this is his best chance to do so.

http://www.nypost.com/seven/10252008/posto...5233.htm?page=0