Election Fraud for Fun & Profit, Part Deux

The story of the optical scan discrepancy has started to circulate around the blogosphere and even made a brief appearance on MSNBC, but is otherwise MIA on the non-Internet media. At the very least, we deserve an investigation into why optical scan ballots produced such an overwhelmingly Republican result in counties with overwhelmingly high numbers of registered Demos.

A thread at Princeton discusses the scan results but dismisses them with a cursory sneer. That's just the nature of Dixiecrats, they say -- they've been registered as Demos since Claude Pepper was a pup, but they vote for God & Guns and against Gays. Smaller towns just happen to use optical scan equipment instead of touchscreens, so that's purely coincidence. Some contributors to the thread note that most of the changes were in small counties that were Bush strongholds in 2000 as well. OK, fine. But there are a few problems with this explanation:

1.) Nationally, voters stayed faithful to their party registration 9 out of 10 times, according to exit polls. For these results to be accurate, some of these counties would have to vote against party affilation by a rate of 60 to 90%. Is it really plausible to assume Florida is so radically different than the rest of the nation? (Yes, they did produce Anita Bryant and they do name freeways after professional football coaches, but come on now.)

2) It's not just small towns. Duval County, with 378,000 votes cast, saw a surge of 56% for Republican votes, and a drop of 9% off expected Democratic results. This is not, as some on that thread suggest, within the range of the touchscreen machines. With touchscreen machines, vote totals rose for both sides in almost every precinct, which is what you'd expect with a massive turnout.

3) Yes, the results do track with Bush v. Gore in 2000. But maybe they were cheating then, too. (In fact, we know they were cheating elsewhere; maybe this form of cheating got lost in the noise).

There is a simple solution to all this. Manually recount vote samples in the counties in question, as Avi Rubin suggested to me when I emailed him about it. That will tell you very quickly whether the scan results jibe with actual votes.

Here's the problem: Florida law requires that recounts be made (wait for it) by the same optical scan machines that ran the faulty numbers in the first place! If the two sets of numbers match (and why wouldn't they?) the recount is over. If they don't match, the scans are run again through the machines. If there are still errors, then a manual recount may take place (though exactly who makes this decision and how they make it isn't spelled out in the recount procedures the Florida Division of Elections has posted on its site.)

To quote Avi Rubin,

"Requiring that they be recounted on the same machines is brain dead. If the machines are faulty or rigged, the same incorrect result will be reached. The whole point of having the paper ballots is to do an independent recount that checks for problems with the original count."

Brain-dead elections in Florida. Gee, who would have expected that?

In other news: The flap about the "extra" 200,000 votes seems to have disappeared. Why? The Florida elections board posted new numbers about voter turnout that now absorb the votes which were unaccounted for yesterday. Is this just a clerical error or yet another fiendish subplot of this conspiracy story? Tune in tomorrow for another chilling installment of As the Scan Turns....