Nate Silver's numbers give me a lot of hope for today.

Nate Silver predicted a popular vote win of 6.1 points for Obama in 2008. He actually won by 7.2. This year, Silver predicts a 2.5 point win and it's very possible this is also an underestimation as you can't really quantify the strong Obama ground game into poll analysis. If this is a free and fair election, we have a strong advantage. Even a 2.5 point advantage would be huge for us, because Romney would probably a need a 1-1.5 point advantage to have a shot at the electoral college because his popular vote numbers by and large come from heavily Republican states, he does not have the numbers in the swing states. So even if Silver is OVERESTIMATING Obama by 1.1 points this time, Romney does not have much hope.

A poll aggregation has a much lower margin of error than the regular polls that are limited to 1000 respondents or so. So as long as his model is good (and was it ever in 2008 and 2010), then we should be VERY confident.