[counterpunch]

Exactly how they are doing it, I can't figure out, because they have the right DRI numbers from 2000 and their gender numbers follow the 2000 exit polling data. Other pollsters are using the same 2000 data and getting entirely different results.

I would suspect targeted polling. What is their area distribution like? Are they targeting big cities and small towns within the margin of voter turnout? What about inner city vs. suburbs? Or perhaps they're just making them up as they go. There is more than one way to skin a cat.

[GAGOPSWEEPTOVICTORY]

Unlike SUSA, they don't release their methodology. They did release party affiliation as well as gender, and I think racial breakdown, but didn't release much else.

I think it's either area distribution or just how they push a likely voter. One or the other.

[Owen]

Folks,

Ohio is as segmented as one state vs another. For example, Indiana and Illinois. Next to each other but philosophically on different planets. There is no law that says a philosophical schism has to only take place across state borders. It can be within the state and that's what Ohio has.

Our Ohio Freepers live in the conservative areas, most likely. They can't imagine Bush losing, from what they see. But they don't see Cleveland and that is the explanation.

Believe the polls. They have a margin of error so if Bush is reading ahead 3 pts in several polls, there HAS to be one now and then that says it is tied. The margin is way too close for comfort. Ohio lost EVs this past census because people left the state. They may have been responsible Republicans who went elsewhere for work.

We have a major fight on our hands in Ohio, and it probably decides the election. Everyone nearby MUST pour your efforts into it.