Pennsylvania - 20
There is no question that the race in Pennsylvania is a lot closer than it was just a few weeks ago. There is no question that Romney is seriously contesting the state. The majority of recent polls show Obama below 50%. But this is probably an instance of too little, too late. Pennsylvania has not gone Republican for president since 1988. That's almost a quarter century of voting for Democrats. Obama won the state by 11% in 2008. He won't win by nearly that much this time, but he should still win. (Obama by 4%)
Ohio - 18
We've been hearing for months how it's all going to come down to Ohio, and guess what? It's all going to come down to Ohio.
Obama won Ohio by 5% in 2008, or by a margin of about 262,000 votes. (Keep that number in mind.) Exit polls show that there was a D+8 advantage in turnout four years ago, and Independents (about 30% of the electorate) went for Obama by eight points.
In 2004, as you probably recall, Bush narrowly defeated Kerry in Ohio by two percent, or a margin of about 119,000 votes. Exit polls showed Republicans had a five point turnout advantage, which was important because Independent voters (25% of the electorate in 2004) went for Kerry by almost 20 points. Republican turnout was almost certainly boosted by Ohio State Issue 1, a ballot measure to make it unconstitutional to perform or recognize same sex marriages in the state.
So it's fair to say that there will be no R+5 turnout this election. But then, there very likely won't be a D+8 turnout, either.
On to the polls. Romney has had the lead in exactly one poll (Rasmussen Reports, by two points) since early October. He has been tied in five others (two of which were also by Rasmussen Reports). In the polls in which he is trailing (which is most of them), the margin has been anywhere from one point to six points.
Let's take a look at the most recent poll from Marist. It has Obama leading Romney 51-45. The sample is D+9. Ridiculous.
The Columbus Dispatch has a poll showing Obama up 50-48. It has a D+3 sample size (possible). Independents favor Obama by 10 percent. That seems unlikely. Are Independent voters really more supportive of Obama now than they were in 2008?
The University of Cincinnati poll shows Obama leading 48-46. It has a D+1 sample (possible). Obama has a whopping (and difficult to believe) 14% lead among Independents.
Gravis Marketing shows Obama leading 50-49. It has a D+8 skew (not at all likely). In this poll, Independents support Romney by 12 points. Hmmm.