By 2 points according to RCP. Big deal.
And the RCP AVERAGE incorporates the errors implicit in the polling. So if one poll uses registered voters while another uses likely voters, the inaccuracy of the former still gets averaged in to the latter.
And if SOME polls still persist in going with respondents weighted with a Democrat voter overload (unintentionally or intentionally doesn't matter) then the skewing of those polls' results will ALSO be "incorporated into the "average" outcome.
Ohio and PA and Florida and VA are going to go with Romney.
Watch.
Ohio, quite possibly.
Florida, I think that's a lock at this point.
PA... as much as I'd like to see that, since I live in PA, I am not sure Romney will take PA. RCp has Obama by 4.8 right now, and I don't think that is a number that can be overcome here in PA.
I don't buy the polls. An "average" that incorporates a number of polls that have (for whatever reasons) methodological errors that skew the results to an artificially high figure for the Democratic is not an average that is worthy of much regard.
I agree that it is certainly possible for the incumbent to "take" PA. But, even so, I suspect it is much closer in PA than Team Obama fathoms. And even if that turns out to be wrong, I at least hope it causes them some sleepless nights!