The Miami Herald's overvote data provides conclusive evidence of the fraudulent counting of absentee ballots in heavily Republican Escambia County, where George Bush won by a nearly 2 to 1 margin overall, and nearly 3 to 1 margin in the absentee vote.

According to the Orlando Sentinel, Escambia County's Republican dominated Election Canvassing Board stayed up until 2AM on election night, "duplicating" over 2400 absentee ballots that were either overvoted or undervoted. Some ballots were looked at two or three times to make sure that as many absentee votes as possible were recorded.

According to the Sentinel, about 10,000 ballots were duplicated in 26 counties. Escambia County accounted for almost one quarter of those ballots. Those 2400 ballots represented over 11% of all the absentee ballots cast in Escambia County.

But the Miami Herald overvote data shows that Escambia County almost certainly was doing a lot more than duplicating what was really on the ballot. The Herald data contains not one single overvoted ballot in which there were only two marks for President. Of the sixty counties where the absentee overvote could be isolated, Escambia is the ONLY county that had no "double marked" absentee overvoted ballots.

According to Florida law, it is illegal to count a ballot on which a mark is made for more than one named candidate. But of the over 21,000 people who voted by absentee ballot in Escambia County, the county board found not one single double voted ballot for which it could not determine voter intent. Throughout Florida, at least 1 out of every 275 absentee ballots cast was double-voted—and that is AFTER at least 26 counties (most of them precinct scanned counties) duplicated absentee overvoted ballots when voter intent could be determined.

The table below provides an idea of how anomalous the Escambia County results are.. Lake County is a good place to start any comparison, because it had a large number of absentee votes and overvoted ballots, but did not hand count and duplicate any of its overvoted ballots.

When the Miami Herald looked at the Lake County overvotes, they found 3103 total. 1615 of those were ballots with only two marks. The Herald was able to determine voter intent on 953 of those ballots, which means that the Herald could not determine voter intent on at least 662 of the Lake County double marked ballots (662 is a minimum, and assumes that ALL the ballots that the Herald found votes on were from the pool of double voted ballots. However, it is likely that at least some of the multiple marked ballots showed voter intent.)

We see much the same thing when we look at Lake County’s absentee overvote. Out of over 11,600 absentee ballots cast in Lake County, there were 227 total overvotes, of which 125 were double marked. The Herald found 86 valid votes among these absentee overvotes, which means that voter intent could not be discerned on at least 39 of the double marked ballots.

And the numbers for Lake County’s poll cast overvotes are similar as well. 2876 were cast, of which 1386 were double marked; the Herald found 867 with voter intent, meaning at least 623 did not show voter intent.

Like Lake County, Escambia County did not look at the poll cast overvotes. The Herald found 3201 poll cast overvotes, of which 1531 were double marked. The Herald managed to recover 421 of the poll cast overvotes, which means that there were at least 1110 ballots with double marks on which voter intent could not be determined.

But unlike Lake County, Escambia County looked at the absentee overvoted ballots, and duplicated those on which it could discern voter intent. And somehow or other, there was not one ballot with just two marks on it for which the Escambia County Canvassing Board could not determine voter intent. In other words, out of over 21,200 absentee ballots cast in Escambia County, there was not a single ballot with two marks on it that the Canvassing Board could not figure out. This is especially odd, when one considers that the Miami Herald could only determine voter intent on at most, 421 of 1531 double marked ballots cast at the polls.

The table shows that there should have been, at an absolute bare minimum, 100 double marked absentee ballots. Even among those counties which hand-counted and duplicated their ballots, the number of double marked overvoted ballots was at least a third of multiple marked ballots, and Escambia had 296 multiple marked absentee overvotes. And according to the Herald data, the Escambia County Board was able to determine the voter intent on 100% of those ballots. The Herald never even came close to a 100% recovery rate on any group of overvoted ballots.

There are thus only two possibilities—either the data provided by the Herald was completely wrong for Escambia County, or Escambia County committed fraud when it counted its absentee ballots. It does not appear likely that the Herald data is incorrect however (requests for the Herald to verify its Escambia County data have so far met with silence.)

The question of what exactly the Escambia County Board did with the ballots is intriguing, but ultimately irrelevant. Perhaps they erased some marks on some ballots, and added marks on others, then duplicated these "fixed" ballots to get the outcome they desired. Or perhaps they just created a whole pile of phony ballots, and pulled all the double marked ballots out of the "overvoted" pile. The fact is that on the night of November 7^th, there was absolutely no reason for the Escambia County Board to suspect that anyone would ever consider scrutinizing their absentee ballots; as the Miami Herald has stated, such scrutiny is "unprecedented".

The possibility of fraud is further enhanced when one considers other actions of the Escambia County Board. Despite having the capacity to tell voters at the polls whether their ballots had overvotes and/or undervotes on them, Escambia County chose to turn off the "voter protection" feature on its precinct based scanners. The reason given was the cost of providing ballots for people who made mistakes (about 23 cents each). In other words, after spending over a half million dollars to buy machines that would tell voters that they had made a mistake, the county "saved" less than $750 by not telling 3201 people that their ballots had been overvoted. Nevertheless, the county did spend over $550 on duplicate ballots for absentee votes.

And unlike every other Florida County, when Escambia provided the Herald with its overvoted and undervoted ballots, it did not provide them precinct by precinct. Instead it presented them in four groups, two that were ballots cast at the polls, and two others that were absentee ballots. Nor has Escambia County responded to various requests for precinct based breakdowns of its overvoted and undervoted ballots.

There is clearly sufficient evidence to launch a criminal investigation into the way in which ballots were counted in Escambia County. That, however, is a matter for the state Attorney General. But there is also good reason for the federal government to launch its own investigation.

Given that over 18% of Escambia's registered voters are of African American descent, and the 9 to 1 statewide margin provided to Al Gore in the African American community, it is highly likely that a very large percentage of Al Gore's votes in Escambia came from blacks who voted at the polls. And given the significant difference in the percentages cast for Gore at the polls and by absentee ballot, it is also highly likely that there was a significant disparity in the percentage of blacks who went to the polls as opposed to voting absentee, with a far higher percentage of blacks voting at the polls. It is also highly likely that there is an historical pattern of higher percentages of blacks voting at the polls. Consequently, the decision to turn off voter protection at the polls while going to the time and expense of duplicating absentee ballots that were overvoted and undervoted must be seen as an intentional act of discrimination against black voters. Such discrimination would be a violation of the United States Constitution, The Civil Rights Act, and the Voting Rights Act, and requires an immediate federal criminal investigation.

The Herald data suggests that Escambia County was not alone in the fraudulent counting of ballots. Bay County, with nearly 12,300 absentee votes, and only six absentee overvotes (all of them double marked) may well have engaged in fraud as well. (According to the Orlando Sentinel, Bay County duplicated many absentee ballots, and is refusing access to the original ballots). And Santa Rosa County, with only a single overvoted absentee ballot out of over 7600 cast also deserves further scrutiny. On average, at least one in every 109 optically scanned absentee ballot was overvoted (at least one in every 91 poll cast optically scanned ballot was overvoted) which suggests that these counties achieved a "voter intent" recovery rate for there overvoted absentee ballots that was far greater than what could be expected.

code