These two charts are definately worth taking a look at. I am interested in knowing if anyone has confirmed or rebutted this information.
The first is a chart comparing the exit poll results versus official results in states that used paper ballots vs. states that used unverifiable electronic voting machines. If these data sets are true, they indicate that the exit polls were inaccurate in states that used unverifiable electronic voting machines (namely Florida and Ohio) but mostly accurate in states that used old-fashoned paper ballots.
The second is a chart comparing registration by party with the election night vote tallies from several counties in Florida.
These data underscore the worries I have previously expressed.