I just completed my absentee ballot, so I'm immune to further campaign ads or anything else that might change in the next week. My election is over. And I'm far from alone. More than 14 million people have already taken advantage of voting by mail, or in-person early voting. So when the media reports that there is an election coming up next week, what they often fail to note is that this election is already in full swing. In Florida, for example, more than 2 million people have already voted either by absentee ballot, or in-person early voting, and a slight majority of those early ballots have come from Democrats. (That is significant because traditionally absentee ballots tend to skew Republican.) In Ohio, the majority of ballots cast so far have come from precincts that President Obama won in 2008. In Iowa, North Carolina and other battleground states, the president is ahead by every measure in early voting.
Whatever the campaigns do in the next few days, whatever happens in the news, whatever polling trends show, all of that is meaningless to people who have already voted. The poll that counts is already over for us. From all reports, that poll is going quite well for President Obama.
The idea that these early voting trends are going to be counter-balanced by some kind of Romney surge on election day seems implausible. What would cause such a shift? The devastating storm on the east coast is only going to slow down the pattern of early voting for a few days, but the president's strong response is likely to drive wavering voters into his camp. We can expect the advantage that the Obama campaign retains in getting out the early vote to continue through "election day." The campaign's task of identifying supporters who have yet to vote, and pestering them to get to the polls, will get easier as time passes.
I'm not an expert on polling, but it seems to me that every poll taken over the next week might suffer from the possibility of taking an unrepresentative sample. The inclinations of voters who are sampled in the coming days become less and less relevant every day, as those prospective voters represent a dwindling share of the actual electorate. What I think that means is that if the polls show a trend towards Romney in the next week, that trend is less important than it might be otherwise, because early voting has already built in an advantage for the president. And if the polls show a trend towards Obama, that will seal his victory, because those who have yet to vote are more likely to break towards the president.
A reminder, however. People who haven't voted yet still need to vote!