Reply #32: From the exit poll data

[Febble]

I've outlined this recently here:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x435991#436807

but also here:

http://journals.democraticunderground.com/Febble/3

And Mark Lindeman has an account of the same analysis here:

http://inside.bard.edu/~lindeman/slides.html

The short version is:

Fraud would be both in reflected in better performance by Bush and by redshift in exit poll precincts. The extent to which Bush's performance correlates with redshift should give you an estimate of the degree to which it was due to fraud.

And it doesn't correlate at all. The full story is fairly complex, but the short story is that trying to squeeze even hundreds of thousands, let alone millions, of stolen vote into that plot requires heroic assumptions. The more parsimonious explanation is that fraud was not on that scale.