[shoedog]

I am not buying all the early voting results, First it was Obama blowing out numbers, now it’s Romney, at the end of the day it is a survey of people who could be blowing smoke, since nobody that I am aware of can share actual results of early votes until election night count me skeptical.

[LS]

No, no one is sure. But we have people at election hq in Dayton every day watching. We've seen a couple of things.

1) The Obama voters of 08 aren't there. This would be busloads of blacks, inner city types. They come in in gaggles. No one is seeing this. A week ago, Obama was here in Dayton, and the Ds had 27 busloads ready to take people to early vote. Ds had three floors ready with tables and people to help them fill out the forms. They expected THOUSANDS . . . got 300.

2) I am told that the Obama voters who ARE coming in are regular "election day" voters, so that the Ds are cannibalizing their election day turnout.

3) Many of the early voters are Rs. My former chairman said it appeared to be 10-15% of all voters this time, as opposed to only 5% early voting Rs last time. But Romney did NOT try to get out "regular" voters to early vote, but rather went after "low propensity voters."

4) I was a little concerned about the Montgomery Co. number, which was up 4000 from 08, but it appears we have a lot of indies. If you assume the indies break 6% to Romney (not much according to most polls), then I put the Montgomery Co. early voter % at 56-44%D. At 12%, that is eight points below what Obama won in 08, when he won the county by about four. In other words, Romney is up, perhaps four, perhaps even more if the indies break harder.

Caveat: this is only one county.

[kabar]

You can estimate the numbers fairly easily. In states that have party affiliation, you can look at who voted early and use those numbers plus independents. In states that don't have party affiliation, you can look at past voting in primaries and other races and estimate from there.

When someone votes early, that is reflected in the poll listing or books that election officers use to see if a person is registered and if they have voted. These listings are updated daily and given to the parties, which use them to identify voters who haven't voted yet and get them out to the polls.

[Windflier]

First it was Obama blowing out numbers, now it’s Romney, at the end of the day it is a survey of people who could be blowing smoke, since nobody that I am aware of can share actual results of early votes until election night. Count me skeptical.

[a fool in paradise]

There were any pollsters asking voters today “who they voted for” so unless the polling agencies are claiming that they have a “good” exit poll sample or have hacked the computers to tally votes, they have nothing but a guess at this point.