Sean Trende has an interesting take on the pull of “gravity” in the Presidential race. Take home: Unless some kind of shock happens (like the 47% speech), gravity has been pulling Obama down such that the election is going to be extremely close and could even favor Romney slightly. This would seem to be consistent with certain political science models such as this one that focus on macro variables rather than horse race events (though some of the macro models favor the President too). His graphic on the first page suggests, though, that some kind of shock being introduced is a fairly regular event – and thus the timing of the next shock and the extent of its effect is going to be critical.
And then there is this which seems to be the only thing that matters – and which favors Obama quite a bit as things now stand: