The odds of an electoral vote/popular vote mismatch probably went up
I would guess that the aftermath of Sandy could somewhat reduce vote totals in a very blue swath of territory (MD, NJ, Eastern PA, NY City area) without affecting the electoral outcome one way or another.
That is not to suggest that such a mismatch is likely. It is likli-er. (2% would be twice as likely as 1%, though both would be negligible.)
Were I betting based on Sandy effects, I would have taken Obama to win the election and hedged that bet with a bet on Romney to win the popular vote. The two would probably cancel each other out, but the odds of winning both would be better than the odds of losing both.
Another possible mis-match effect—payroll data versus unemployment % in Friday's jobs report. Sandy presumably had a deleterious effect on jobs. (Losing 2-3 days matters in a monthly number) But perhaps no such effect on the household survey. (Maybe more employed people were home to respond?)
= new reply since forum marked as read
