Reply #13: I hate the way he portrays the different studies. He makes it into some [View All]

[milkyway]

kind of game of two equal sides that no layman could possibly figure out.

But that's not true. The CalTech fake study was crap. They used some of the later "weighted" numbers that were adjusted to more closely reflect the actual vote count. Then they use these numbers to show the exit polls were close to the actual vote count. WTF? Of course they're close. They were changed to make them close.

They didn't even explain where they got their exit data from, and had no discussion of the weighting that was done to the late exit poll numbers. They also had the nerve to complain about another analysis done by somebody else that they didn't say where their data came from.

Near the end, they had a totally bizarre statement that you can't be sure that exit polls are accurate unless you polled every single voter. WTF? That's what polling and statistical sampling are all about. It was a fraudulent non-study.

Also, the study by Steve Freeman at UofPenn was a very thorough study, backed up by documented evidence. It was about 28 pages long, compared to the junk study by CalTech that was 2 pages long (if I'm remembering correctly).

But just when I'm feeling pissy toward Olbermann, he more than redeems himself with the coverage of Ohio. And he doesn't just pass along what's being reported--he actually makes some very astute points of his own.

He speaks of some counties that bush won as having a lot of uncounted undervotes. This seems to refute the charge that the machines with a lot of undervotes were biased against Kerry. But then Keith says this: "or did the precincts wind up voting for Bush because more than a quarter of the ballots had no valid presidential vote?" (In other words, bush won those precincts because Kerry votes were deliberately undercounted.)

Great point, Keith. Now get your crack staff on this and see if you can answer this question.