On today's show, MSNBC's Keith Olbermann promulgated the myth that somehow results in Florida are tainted because certain counties that have more than 2-to-1 Democrat-to-Republican registrations overwhelmingly voted for Bush. Counties he cited on tonight's show were: Baker, Lafayette, Liberty, and some others. He repeats this on his blog, but with less detail here (the transcript of the show is not yet available. Will post as soon as it will be):
... huge margins for Bush in Florida counties in which registered Democrats outnumber registered Republicans 2-1, places where the optical scanning of precinct totals seems to have turned results from perfect matches for the pro-Kerry exit poll data, to Bush sweeps.
So, I decided to go to the Florida Secretary of State's office webpage and look at the results for Lafayette, Baker, and Liberty, the three counties whose names I could catch, and see how they voted in 2000 and in 1996. Well, quite a fascinating result:
Election of 2000
Baker voted 5610-2392 for Bush against Gore.
Lafayette voted 1670-789 for Bush against Gore.
Liberty voted 1317-1017 for Bush against Gore.
Election of 1996
Baker voted 3684-2273 for Dole against Clinton
Lafayette voted 1126-899 for Dole against Clinton
Liberty voted 913-868 for Dole against Clinton
Game. Set. Match, Keith. I loved you on ESPN, but you should stick to what you're good at: sports. The reason these counties have such high Democratic registrations is because they likely still have traditional Southern loyalties to the Democrats, but being in Northern Florida and the Panhandle, they are all very very conservative. No wonder, they even voted for Dole! As soon as the transcript of his show is up, I will check the other counties for which he peddles the conspiracy theory. (UPDATE: I have a post above that considers the 28 counties where the disparity between Bush's 2004 totals and Republicans' share of registrations is the highest... Let's just say that Olbermann has a whole lot of nothing).
Why am I not surprised that the wacky left couldn't accept the results of this election?
UPDATE: Welcome Volokh Conspiracy and The Corner readers! Thanks to Prof. Volokh and Jon Adler (who is a former editor of The Yale Free Press) for linking. So, check it out and stay for a while. Also, see this post I made, doing a full county-by-county analysis of these "disloyal" counties, showing their totals for the 2000 and 1996 election. Also, check out the articles that have appeared in our print edition.
Yes, I do know I misspelled Olbermann's name in the title, but Blogger puts the title in the URL for the link and I noticed the problem only after people started linking to me. One thing I realized I didn't touch enough on is how this seems to demonstrate my previous point that the Left hates Bush no matter what. Even with this election, which was quite clean and free of controversy, they have to invent controversy because they cannot believe that they could lose to this simpleton. So much for uniting the country and healing the wounds.
UPDATE: Keith Olbermann still seems to stick to his story, while conceding that Bush won most of the counties Olbermann mentioned, in 2000. Olbermann just says that he mentioned mostly panhandle counties and should have presented a better mix of the other 24. I'll try to have a more thorough analysis of the other cases where the voter "disloyalty" phenomenon occurred, later. But on face, this seems like Obermann is seriously stretching the bounds of plausibility.
UPDATE: The myth is repeated here for Baker County. It also claims that a ballot initiative in favor of raising the minimum wage in Florida passed with more than 70% of the vote, while Kerry only got 48% (Olbermann mentioned this point on his show tonight as well). Does that mean that since the initiative to define marriage as that between a man and a woman was supported overwhelmingly in Oregon and Bush lost overwhelmingly there, that the election in Oregon was stolen? My understanding is that Southern Democrats tend to be economically liberal but socially conservative. They might vote for some populist measures, but on social issues and national security they will vote for the Republicans. Another interesting paradox of that sort is the success of trial lawyers like Dickie Scruggs in Mississippi, where they find some small-town jury and sue a big corporation in that small town, and almost always win because they can frame the debate in terms of David vs. Golliath. Mississippi overwhelmingly votes for Republicans for President, who tend to support tort reform, but still hold lawyers like Dickie Scruggs in high regard. One final point is that states like Indiana and North Carolina have Democratic legislatures, but also overwhelmingly vote Republican for President. Louisiana hasn't had a Republican Senator since Reconstruction (until this year, when David Vitter won to replace the retiring John Breaux), but also always votes Republican for President.
UPDATE: Similar, albeit more "scientific"-looking claims are made here. Still bunk because the analysis entirely ignores the vote from 2000 and 1996 in those counties. They compute an expected number of votes that each candidate should have received in various counties. Their expected number votes is entirely based on the voter registration by party in that county:
EXPECTED_VOTES REP = the percentage of registered REP * the total number of voters who voted in each county on Tuesday.
EXPECTED votes would normally vary from the ACTUAL votes due to increased voter turnout by one party, Independents voting REP or DEM or other factors. What seems very odd in these numbers is that the increase in ACTUAL votes from EXPECTED votes has a striking pattern of being so much higher for REPs than that for DEMs in counties using optical scan voting machines, even when smaller counties are excluded from the analysis.
And would their model of expected votes account for the election results in 2000 and 1996 for Baker, Lafayette, and Liberty? Most definitely not. Any good model ought predict at least past events before it can be used as a candidate for future events!
UPDATE: This is a thread at the Democratic Underground on precisely this issue. One post says:
If they have officials committing fraud, it's likely they had them then too. In fact, we know elections have been rigged in the deep south for generations upon generations.
Proof that the Republicans have ALWAYS won in a heavily Democratic area could perhaps be a strong case FOR fraud, and in fact that seems to be the case, prima facie. Your evidence may support the fraud allegations, in other words.
That doesn't make sense. I just checked on the Florida Department of State page and for example, Baker County voted for Jimmy Carter (a Southern Democrat) over Ronald Reagan in 1980 (the last Presidential Election for which data is available there) by a margin of 2606-2271. So, it seems that this long tradition of fraud for generations and generations doesn't make much sense. This is especially true of the South, which only until the last 25-30 years, had voted predominantly for Democrats for President. So, sorry, this won't cut it. I suppose I shouldn't be surprised by stupidity being promoted on the Democratic Underground site.