Statistical projection of already-completed manual recounts would increase Bush's statewide margin of victory by 176 votes. Counties selected for recounts by Gore had fewer so-called "anomalies" than many other counties.
Contrary to some public accusations, the 2000 presidential election is not being stolen by biased election judges who are fraudulently counting more votes for Al Gore than for George Bush. That doesn't mean that counting every mark on a ballot is a sensible approach; the process in fact does create additional votes with little certainty that they are real. But the real theft lies simply in the fact that complete manual recounts have been demanded in just four heavily Democratic counties in Florida.
Two of those counties, Broward and Volusia, have now finished complete manual recounts. As a result of these recounts, Gore's total vote count in those two counties increased from 483,581 to 485,073. That increase of 1,492 votes represents a 0.30853 percent increase in his total. Bush's total count increased from 259,491 to 260,307; his increase of 816 votes represents an increase of 0.31446 percent. In other words, the percentage increase was almost exactly the same for each candidate, differing by less than one ten thousandth of one percent -- an incredibly miniscule difference.
In spite of the fact that both candidates percentage gains were almost exactly the same (in fact, Gore's was ever so slightly smaller), the recount yielded Gore a net gain of 676 votes. That's because the recount, while increasing each candidate's total votes proportionately, took place only in two counties that Gore carried decisively (by 65 percent).
Manual recounts by their very nature tend to increase the total vote count. That's because human beings can count (or attempt to count) ballots that machines cannot. They can count small pencil marks, scratches, dimples, impressions, nicks, and other markings that machines cannot.
However, as the results in Broward and Volusia counties have already indicated, these increases add votes to all candidates proportionately. Further, the partial results of the halted manual recounts in Miami-Dade and Palm Beach counties indicate similar results.
So far, however, the votes of one candidate have been recounted disproportionately (485,073 for Gore versus 260,307 for Bush). Furthermore, county-by-county totals indicate that the choice of counties to recount was not based on so-called undervote "anomalies," as claimed by Gore.
Of the 36 Florida counties that use optical-scan ballots, 21 had higher undervote rates than Volusia county's rate of 0.29 percent before its manual recount and 13 had undervote rates more than twice that of Volusia county. In fact, two optical-scan counties had undervote rates more than eleven times greater than Volusia county (and also greater than Palm Beach). Of the 18 counties that use punch-card ballots, 10 (excluding Miami-Date and Palm Beach) had undervote rates greater than Broward county's rate of 1.13 percent, five (excluding Palm Beach) had undervote rates greater than Miami-Dade county's rate of 1.64 percent, and two had undervote rates greater than Palm Beach county's rate of 2.29 percent.
In total, four Florida counties, all carried by Bush, had undervote rates greater than those of much-publicized Palm Beach county. Even the infamous 19,120 Palm Beach overvotes -- allegedly cast for Gore and Buchanan due to poor ballot design -- were exceeded in percentage terms in no fewer than 16 other counties, 15 of which Bush carried.
If the manual recounts in Broward and Volusia counties are accepted, and the percentage increases for each candidate in those two counties are projected statewide in Florida (instead of counting them in isolation), Bush's total vote total of 2,911,729 after the first machine recount (excluding Volusia County's manual recount totals) would increase by 9,156 votes to 2,920,773 while Gore's total of 2,910,701 would increase by 8,981 votes to 2,919,742. This would give Bush a new lead of 1,204 votes. That would be a change of just 176 votes from the 1,028-vote difference after the first recount (excluding Volusia County's totals).
The plain reality is that, if every ballot in Florida were to be recounted -- rather than just the votes in four selected counties -- there would be no change in the statewide election results. However, it does not make sense for Bush to undertake a statewide recount effort that would leave the presidential election hanging in the balance for many additional weeks, extending the contentious debate of the last three weeks, if its likely result is to increase the margin of victory by just 176 votes.
Significantly, Gore has never sought such a statewide recount, his public statements notwithstanding. In fact, he has refused to do so on three separate occasions. He did not request a statewide recount within 72 hours after the original county certifications, as required by state law. Later, during the November 20 hearing before the Florida Supreme Count, his attorneys were asked if they were requesting the court to order a statewide recount. Gore attorney David Boies replied, "We are not urging that upon the court." In its decision, the court pointedly noted, "At oral argument, we inquired as to whether the presidential candidates were interested in our consideration of a reopening of the opportunity to request recounts in any additional counties. Neither candidate requested such an opportunity." Finally, his November 30 appeal to the Florida Supreme Court asked the court to order that "only the ballots contested in the Complaint be examined and counted."
If Gore doesn't want to "count every vote" in the state of Florida -- as he repeatedly states, but only the votes in four selected, heavily Democratic counties, then what is his true goal? Sadly, it seems that he seeks victory at any price. In so doing, he cheapens our electoral process, undermines the rule of law, and threatens our democracy.
Jerome F. Winzig is a freelance technical writer in Minneapolis, Minnesota. He wrote this article for the Northern City Journal.