The ABCs of Polling
by Tom Donelson on September 29, 2012 at 12:19 PM
For those who read polls on a daily basis and see an Obama sweep coming, let me explain a few pointers about the polls you read.
First, most of the polls are skewed to Obama. How, you may ask? By oversampling Democrats. Most pollsters are basing their polling data upon a model that's equal to or exceeds the 2008 race. Democrats came out in full force in 2008 and had a seven point advantage, reflected in Obama's margin. The question is will this actually happen again? If it doesn’t, Obama will lose. Those pollsters argue in favor of this because of a higher minority turnout and demographics that favor Democrats. And this is not an unreasonable argument.
However there are factors that go against this. In 2010, Republicans exceeded their 2008 percentages. In an economy that sucks and a Middle East that is imploding, the odds that Democrats repeat their 2008 performance is nonsense. Jay Cost of Weekly Standard has made the case that the polls, including those in battleground states, are skewed in favor of Obama by 3 to 4 points. While it is possible that pollsters are right that Democrat turnout will match their 2008, the odds are against it if Jay Cost believes that turnout may be in between 2004 and 2008. This would resemble the average turnout over the past four decades and points to a close race.
Second, How many of these voters are willing to change their minds? In reviewing polls from various battleground states, I have noticed that many voters are not only undecided but are willing to change their vote if the situation warrants it. I have seen as much as 20% who are either undecided or willing to change. These are the unpredictable aspects that pollsters can’t truly measure. Conventional wisdom has very few ready to change and the undecided shrinking. I disagree, and there is enough data to show this to be the case. The question is whether this benefits the incumbent or challenger.
In 1980, Carter had 45% to 50% throughout the fall of 1980, and there were times his polls numbers were similar to what we are seeing now. He ended up with 42% and Reagan significantly outperformed every poll number he had before the election. Many of those undecided and those willing to change went for Reagan. So can it happen again? Yes.
This brings me to final point, the election is close and within one or two points. Obama, just like Carter, has the media on his side and running interference. Let's face it, if you read the New York Times, you would be hardly aware that Obama's policy has collapsed in the Middle East and you read the economy is humming on all cycles. The New York Times has done a pretty good job of hiding what is really happening, and essentially the front page mimics Democratic talking points. Reagan had the ability to go over the top of media and in the final debate, he showed himself Presidential and hardly the crazy old coot who was just anxious to let loose the nukes on Russia. There is the problem that Romney is no Reagan when it comes to making his case. The question is will the reality of what is happening, in the Middle East, Middle Class incomes dropping, America's standing in the world dropping, and a President out of his league pop through the media's pandering for Obama?
It is up to Romney to make his case, and for those of us who understand this country is in deep trouble to outwork our opponent. This race is far from over, and we are not dealing with a great Philosopher king but an overrated talent who is not as smart as he or the media thinks he is and an America that is imploding. Will voters overlook the media message and see what is truly in front of them? And can Romney point the way effectively? This won’t be 2008, but it could be 1980.