I am, like a lot of Kossacks, kind of getting tired of repeatedly being barraged by exposure to the US Count Votes article claiming that a non-fraudulent explanation for the discrepancy between exit polls and election results is highly unlikely, some even going so far to imply Kos is burying the story. Let's talk for a moment about probability.
What really gets me about this isn't the fancy statistics, it's the logical implication of assuming that there was, in fact, significant fraud that was outcome-determinative. This election hinged on Ohio, which we lost (according to the official count) by hundreds of thousands of votes.
A shift in votes of this magnitude - and yet small enough on the county-level to raise obvious suspicions - would require the willing cooperation of tens if not hundreds of people all across the state of Ohio, from the state government down. What's being posited here, I believe, is a conspiracy on a huge scale.
I'm not saying there was not some minor fraud in Ohio; we know for a fact that certain counties did strange things, and my gut tells me that both sides probably cheated a little where they could get away with it. In Texas, this has been the rule, not the exception.
But to make up three percent gap, which again is hundreds of thousands of votes, seems to me to be almost impossible.
In examining what would be required in pulling that off, I am reminded of the debates I have had with my father about the Apollo moon landings. A few years ago, after watching a special aired by FOX alleging that the moon landing was a hoax, my father was convinced that we had never landed on the moon. Even after explaining to him that all of the alleged evidence of hoax (no shadows! radiation!) were all, actually, scientifically explainable (as indeed, I think the exit poll discrepancy is), I still couldn't convince him. Which led me to the main issue:
Do you think a bureaucracy with thousands of people working in it could keep a secret of that magnitude for 30 years? It is simply contrary to all known laws of human behavior.
This is why I am skeptical of the vote fraud theory and you should be too. Moreover, I am skeptical because people I respect when it comes to analysing these issues, such as Ruy Teixeira and Mystery Pollster, are dismissive of these claims.
And before you say Diebold - most of Ohio was using traditional voting methods on Election Day.