First it was Obama had a fantastic, excellent debate. Then it was all part of a master jujitsu chess plan. Now the pollsters are corrupt.
Seriously? Really?
Look, there are pollsters out there who are crap. Rasmussen comes to mind. We Ask America. This Gravis whatever it is. But there are pollsters who have spent years building up a solid reputation for thoroughness and accuracy. You cant trash a good pollster when said pollster says something you don't want to hear. Nor can you tout a crap pollster when it shows something you like.
This is nutcase territory. What's next? One of these "unskewed polls" websites where you guys are going to re-weight the poll to fit your expectations of what the electorate ought to look like?
I went back and compared the last Pew poll which had Obama up 8 with this one. Guess what I found. The changes in demographics were almost nil. Minor. I didn't hear you all trashing the poll last month.
Lets look at what REALLY happened in the pew poll. Look:
The last Pew Poll ended up being 37% D, 31% R, 31%. This one has 36% R, 33% D, 30% I. I didn't hear at any point folks here saying 'the Pew poll is skewed towards Democrats' but there were plenty of nuts on Redstate saying it. We were laughing at them them.
The truth of the matter is that Pew does not weight its polls by its own perceptions of Party ID ought to be or what it thinks the electorate will look like in November. You don't know that no matter what you think you know. You simply know past events which although is the greatest predictor of future behavior, certainly isn't perfect. So, people self identify their party preference and this is a factor that will vary. It will vary on moods of the electorate. It will vary depending on the time of day the phonecalls are made. But even this change in party ID doesn't explain the sharp shift in the Pew Poll. What does? Well, look here:
Obama was 49 to 46 among white women in the previous poll. He's losing 57 to 38 among white women now. And you can see in the details that most of the damage was among white women with less than a college education. There Romney was leading 49 to 45. Now he's leading 63 to 31.
These are the low information voters. These are the white ladies who work and don't follow politics online because they probably don't sit at computers while doing their jobs. The probably don't watch the news very often because after busting ass all day and taking care of kids, who wants to? The debate watch number was huge...60+ million. More than any debate in the 2008 cycle. It isn't unreasonable to say that for millions of these women they got their first look at Mitt Romney outside of ads and media coverage. And he made the best of that opportunity. Meanwhile, our guy was somewhere else so there was nobody to call Mitt Romney on his lies and bullshit.
There is probably some noise and variance in there, but for the most part, what you see reflected in the polls is a shift among non college educated white women as result of the debate. Obama's number among college educated white women was still a solid majority. So, there is one part of Obama's coalition of minorities and white women that is "swingy."
What that tells me is that these ladies not only want to hear about policy, they want to see a fighter who is going to go to the mat for them. These women are the ones who are going to have to rely on Social Security and Medicare since a huge number are unmarried. They don't need to see a President Obama nodding his head in agreement about cutting Social Security. These are the ladies that need WIC, Food Stamps, Medicare, Medicaid, Social Security, and they want someone to fight for these things.
Because otherwise, may as well try out the new guy.
For all you folks trashing the Pew poll, I had better not ever see you tout another Pew poll. I don't care how much the numbers improve, you and Pew are done.