Today’s Courier Herald Column:
This is the column where I’m supposed to tell you what’s going to happen Tuesday night. Except I can’t, and I’m not sure who can. We’ll know soon enough anyway. At least we can hope so. While there are predictions for landslides for both Presidential candidates, there exists the possibility that one or more close swing states will push the verdict on this election into counting provisional ballots and possibly, recounts.
Polling still suggests that President Obama has a slight lead in both the popular vote as well as most swing states. As of early Monday morning, the Real Clear Politics average of polls has President Obama narrowly winning the popular vote by three tenths of one percent, while having a landslide in the Electoral College. With a late move shifting Virginia from Romney’s camp to that of the President, they give President Obama a 303 to 235 decisive victory.
All hope is not lost for Republicans, however. Their same poll averages still lists 146 electoral votes as toss-ups. Many of the polls showing the President with a commanding lead also assume that Democrats will vote with the same intensity or higher than they did in the landslide election of 2008. Crosstabs within these polls however generally show Romney winning Independents handily and Republicans showing much more enthusiasm for Romney (or for voting against the President) than they did in voting for John McCain four years ago.
The result is an electoral playing field that is wider than usual. Romney’s team has decided to make a late play for the usual Democratic bastion of Pennsylvania. Some polls indicate support for the President is soft there among Democrats, while Republicans have made up ground in Congressional races within the state recently. The Romney campaign has spent millions in a late ad buy and gave crucial time during the last 72 hours of the campaign for a personal visit.
The late action in Pennsylvania by Romney could suggest that other states are sewn up, but most likely acknowledges that other swing states are already saturated with ad buys. It could also mean that Romney sees that the President’s firewall in Ohio is holding and he needs a plan B to get 270 electoral votes.
The fact that the Romney is in Florida the day before the vote suggests that the states that appear to be breaking for Romney are not being taken for granted. The fact that former President Bill Clinton will be making 4 stops in Pennsylvania on Monday suggests the President’s campaign isn’t sure it has Pennsylvania locked up, either.
For the first time in a while, the Presidential race will come down to who shows up with the most intensity at the polls. Neither side is projecting overwhelming confidence in victory.
The result of the national election will also tell us about the accuracy of polling in the modern, wireless era. Fewer and fewer of us maintain land line telephones any more, and even more of us have programmed ourselves to send any unknown caller into voicemail. It would seem that pollsters are having to glean information from a smaller and smaller segment of the population – one that is becoming more and more self-selecting.
While the polls and their accuracy are something that the inside baseball crowd can argue in perpetuity, there is a real issue with their accuracy with respect to policy and the average citizen. Many news outlets have become accustomed to reporting both issues of the day and early candidate viability based on public opinion polls. What we see reported is often determined by issues polling, along with implied editorializing of the news based on how many of those polled agree or disagree with or approve of the issue or situation.
We’ll know who our President is soon enough, but we’ll also learn a lot about those who poll us. If they’re right, it will likely be business as usual. If they are not, we’ll need to take a long look at how polls are used in areas beyond campaign predictive models. As of now, however, the only poll that matters will be the one counted on Tuesday.