Burnt Orange Report

Exit Polls and Other Data

By Byron LaMasters

It's out all over the place, but since a lot of the blogs are going down due to extraordinarily heavy traffic, I'll post what I've seen throughout the afternoon. Again, I stress that exit polls are a flawed science, and read this disclaimer before you try and draw a lot of conclusions:

Slate and Kos have posted these numbers at 4 PM:

NV CO NC PA OH FL MI NM WI
Kerry 48 46 49 54 50 50 51 50 51
Bush 50 53 51 45 49 49 47 48 46

If they're correct, this election is about what we've thought all along -- Ohio and Florida. If Kerry wins one of two, he wins.

Kos and MyDD have great news leaked from CNN:

Ohio - African American precincts are performing at 106% what we expected, based on historical numbers. Hispanic precincts are at 144% what we expected. Precincts that went for Gore are turning out 8% higher then those that went Bush in 2000. Democratic base precincts are performing 15% higher than GOP base precincts.
Florida - Dem base precincts are performing 14% better than Bush base precincts. In precincts that went for Gore, they are doing 6% better than those that went for Bush. African American precincts at 109%, Hispanic precincts at 106%.

Pennsylvania - African American precincts at 102% of expectations, Hispanics at 136% of expectations. The Gore precincts are doing 4 percent better than bush precincts.

Michigan - Democratic base precincts are 8% better than GOP base states. Gore precincts are 5% better than Bush.

I tend to think that applying actual turnout data to historical patterns is much more significant than exit polls. Seeing these turnout numbers in Democratic precincts gives me much more confidence than exit polls (which can be all all over the place). It just very well may be ACT and the Democratic 527s that will win this election for us.

Looking over at National Review's The Corner Blog - there's a good deal of exit polls leaking out. Speaking of the fickleness of exit polls -- they had Bush up by 8 in Ohio in the first batch of their exit polls, and then one came in with Kerry up 4.

Drudge also has exit polls that show Republicans doing well in Senate races:

Thune +4 (SD)
Castor +3 (FL)
Burr +6 (NC)
Bunning +6 (KT)
Coburn +6 (OK)
Demint +4 (SC)
Salazar +4 (CO)

Another bit of hard data that gives me comfort is the fact that in real numbers from Dixville Notch and Hart's Location in New Hampshire (that have voted and been counted), John Kerry outperformed Al Gore (from those locations in 2000). Matthew Gross analyses it. I had a similar thought this morning (although I didn't post it here) in my comment on Boi From Troy's post on the story.

Here's the trendlines (comparing Bush/Gore in 2000 vs. Bush/Kerry in 2004):

Location: 2000 (B/G) 2004
Dixville Notch, NH: 21/5 19/7
Hart's Location, NH: 17/13 15/15

So Bush's twenty vote margin of 38-18 in these two towns in 2000 was reduced to a twelve vote margin of 34-22. That's a seven point swing towards Kerry as Bush's percentage in the two towns has shrunk from 68% to 61% from 2000 to 2004.

Update: Via Atrios is Zogby's final prediction: Kerry 311, Bush 213

Update: Kos has the 6 PM numbers:

Kerry Bush
PA 53 46
FL 51 49
NC 48 52
OH 51 49
MO 46 54
AR 47 53
MI 51 47
NM 50 49
LA 43 56
CO 48 51
AZ 45 55
MN 54 44
WI 52 47
IA 49 49