That is so out of the realm of rationality that it becomes a problem. I hope all pollsters are seeing Ohio with a 42/34 D:R spread. That means the Republican has to be winning by an entire marathon.
That is just insane. Ohio is always close. A run-away in Ohio is a 4-6 point win by either side. The reason it's a longtime swing state is that the Ds and the Rs balance each other out.
FWIW, whenever I rebalance a poll in Ohio, I use the 37%R/36% D number that Pew, I think, came out with earlier this summer.
That is just insane. Ohio is always close. A run-away in Ohio is a 4-6 point win by either side. The reason it's a longtime swing state is that the Ds and the Rs balance each other out.
I dunno, I'm not so convinced of that. I will grant that I'm NOT a political scientist, so my methodology may be missing something that I'm just not aware of, but when I rebalanced this poll using the Pew 37%R/36%D numbers, and adjusting Independent spread with the lead for Romney presented in the data set, I have the poll at 49.6% Romney, 42.0% Obama.
Further, when I take the usual 3:1 break of undecideds for the challenger, subtracting 2% for third parties, I project a final result of 54.4% for Romney, 43.6% for Obama.
That could just be way, way off because, as I said earlier, I'm not a political scientist, and I may just be bumbling my way around in the dark like some kind of retarded chemist-trying-to-pretend-to-be-a-political-scientist.
Nevertheless, your statement that Ohio is *always* close is not necessarily accurate. In 1988, Bush I beat Dukakis by 10.85%, and in 1984, Reagan beat Mondale by 18.76%.
Now, sure, Romney is no Ronald Reagan. BUT, Romney is pulling in Ronald Reagan-like numbers with white voters, as shown on another thread earlier. The numbers I calculated from the historical trend in GOP capture of white vote versus actual popular vote tallies projects Romney to get around 52-53% of the total popular vote, which is right in line with Gallup's numbers from the past few days, and which is ALSO around what Bush I got in the popular vote in 1988 (whose winning percentage spread in Ohio was ALSO very close to what I've projected in Ohio for Romney).
Again, maybe I'm an idiot, and I certainly grant that this can be the case, but it just seems to me that too many coincidental numbers are falling into line for my analysis to be totally off the beam. LS, any thoughts from the resident PoliSci guru?