I have been looking at the numbers from the polls in the counties in Florida, specifically Alachua County because that's a place, after living there for seventeen years and being politically active, I know something about.
The site where you can see the data yourself is http://ustogether.org/Florida_Election.htm.
The data she presents is for 2004 for all the counties in Florida, devided according to the kinds of tabulating machines they use. Like New Hampshire, Alachua County uses optical scanners from Diebold. If you look at the numbers across the line, you will see that the Democrat won. If you go further down the page and call up the numbers for 2000, you'll see that the Democrat won in 2000 as well. Alachua County has long been a Democratic sea in a conservative part of the state.
From 2000 to 2004 registration in Alachua County increased quite significantly. Republicans added about 7000, Democrats added about 10,500 and Independents increased their numbers by about 8000. About 75% of registeredvoters went to the polls in 2000. This year that increased to 78%. If we assume that 22% of each group stay at home on election day, then you get the number in the ?expected vote? column. If you assume that every single person registered as a Republican went to vote, both this time and last, then the actual number of Republican votes was 2000 and 8000 more than the number of registered voters in 2000 and in 2004. In other words, we have to believe that 6000 more non-Republicans voted for Bush this time than voted for him last time.
At the same time, even though Democratic registrants increased by over 10,000, we are supposed to believe that 9,500 Democrats stayed home this time, just as 14,000 stayed home in 2000. That there were still more Democratic than Republican votes may be accounted for by Independents. Although, if any Republicans stayed home, there weren't a lot of Independent to fill the hole.
Although there is every reason to dismiss Alachua County as unimportant from our perspective because the Democrat carried it, I think it deserves a closer look. How do we account for the fact that Republicans garnered over 16,000 more votes than expected and Democrats only took in 6,000? Where did those extra 16,000 votes come from?
By the way, these totals are not reflected in other county-wide races for the U.S. Senate, State Attorney and Sheriff which run in the 86,000 to 90,000 range. So, a lot of people obviously came out just to vote for the top of the ticket and more selectively for other positions.
So what do I think? Well, I think that just as the DFA blog doesn't like certain words (like Y!go) and rejects them, the program that runs the optical scanner can not only reject certain votes (called spoilage), which nobody seems to pay much attention to, but may also be able to send some votes (maybe every fifth or tenth) for one candidate to another's column. If the percentage is low enough in counties where the opposition is strong, then this leakage is not likely to be noticed. In a tight state-wide race, however, it is almost certain to throw the win to the favored candidate, while in a race where the candidate is so bad that everyone votes against him (I give you Benson in NH), the margin will be large enough so the leakage is insufficient.
Let us hope that Ralph hasn't lost all his marbles and that is what he is setting out to prove by calling for a manual recount in New Hampshire. As a candidate and a loser he has standing to challenge the results. Winners don't. I don't know who else does.
Seems like a good question for the lawyers.
BTW, since Georgia used optical scanners as well, one is led to wonder if that accounted for the unexpected loss by Cleland and Barnes. Sometimes what we don't see is more important than what we do see, especially if what we do see has been structured as a diversion.Posted by Hannah at November 9, 2004 08:27 AM