William Kristol writes that the polls' results seem bogus due to the disingenuous responses to the presidential vote question in number 12:
Looking at these results, one becomes pretty inclined to declare this a rather bad poll. As Kristol rightly points a poll that shows Obama one point behind in Arkansas, when carried Arkansas in 2012 by 24 points. Similarly, the Kentucky sample is +3 Romney when reality was +23. The Louisiana sample is +3 Obama in a state Obama lost by 17, and the North Carolina sample is +7 Obama in a state he lost by 3, he notes. Feel free to dismiss it as "right wing media" but it was noteworthy enough for the BBC to mention it (which is where I read it first)
Nate Cohn writes that this can be explained by the "well-known bias toward the victor in post-election surveys. Respondents who voted for the loser often say that they don’t remember whom they supported, or say they supported someone else". This seems rather far-fetched to me however, considering polls show Romney victories in hypothetical re-votes and that the President is considerably underwater in the opinion polls, particularly in these states. It would seem the results of this poll, then, cannot amount for much.
The presidential vote question was asked of all adults. The 2014 election questions were asked of only registered voters.