The Stealing of the 2000 Presidential Election

I'm currently working with a publisher, Biting Duck Press, to publish a book (title as yet undetermined) on the corruption in our election system. We intend to have it published prior to the November election, and hope that it will help to make Americans more vigilant and concerned about the way our elections are run. I’ve drafted most of the book. I am currently intending to post large portions of it on DU, in the hope of stimulating discussion and obtaining useful feedback. I’ll start with a portion of the first chapter:


THE STEALING OF THE 2000 PRESIDENTIAL ELECTION

Going into Election Day 2000, the Presidential election between Al Gore and George W. Bush was judged to be very close. Three battleground states with early closing times – Florida, Pennsylvania, and Michigan – were polling very close and held the key to victory. Pundits had been saying that if either candidate won all three of them, that would be the ball game.


Election Day 2000

Florida initially looked very promising for Gore. Turnout appeared to be very high, especially in African-American and other Democratic areas. Florida was called for Gore early in the evening, beginning at 7:52 p.m., less than an hour after their polls closed. Then the Pennsylvania polls closed, and the election there was said to be too close to call. Shortly after that, the Michigan polls closed and Michigan was immediately called for Gore. An hour or so later, Pennsylvania was called for Gore. With the polls yet to close in the western states, the election appeared to be virtually over, with a victory for Gore.

But then, in a very rare reversal of a network call, starting at 10:13 p.m., the networks took Florida out of the Gore column and called it “too close to call”. Not long after that, it became evident that the winner of Florida would win the election. Then, at 2:16 a.m. Wednesday morning, the networks began calling Florida for Bush. At 2:30 a.m. Gore phoned Bush to concede Florida and the national election. It was at that point that I went to bed.

At about 3:30 a.m. my wife woke me up to tell me that Florida – and with that the whole election – had been put back in the undecided column. I didn’t believe her. The professionals surely wouldn’t reverse their call on the same state twice in the same election! I went back to sleep.

I woke up later that morning to go to work, looked at the TV, and noticed that Florida was now colored white instead of red on the electoral map. What the hell was going on? I listened to the news reports and discovered that Bush was still ahead in the election by 1,784 votes – about three hundredths of a percent lead. Because of the narrowness of the Bush lead, an automatic recount was proceeding, as required by Florida law. That recount, conducted by the same machines that produced the original vote count with Bush up by about 1,784 votes, ended on Friday, November 10, with Bush leading by about 327 votes according to the Associated Press, with all counties having reported their results. By that time, Gore had already requested a full hand recount of four Democratic counties.


Why couldn’t the TV networks get it right?

The TV networks explained their two bad calls (calling Florida for Gore, reversing that, calling it for Bush, and then reversing that too) with the simple phrase, “bad data”. In other words, they hardly explained it at all. But a consideration of those two bad calls – for Gore, and then for Bush – is very important in providing an understanding of what went wrong in this election. Similarly, it is also very important to understand why the automatic machine recount reduced the Bush margin of victory by about 80%.

First bad call – The butterfly ballot and exit polls deviating from the official vote count
First, let’s consider the first bad call – for Gore. Why was Florida called for Gore so early? The fact is that early election predictions and calls are based on a combination of exit polls and official vote counts. Exit polls measure who the voters think they voted for. Normally voters know who they voted for. But in Palm Beach County, Florida, in 2000, a “butterfly ballot” was used for voting for President. The butterfly ballot was very confusing, as Al Gore’s name was listed next to two third party candidates – Patrick Buchanan and Socialist candidate David McReynolds – on the adjacent page, making it difficult to tell which hole punches corresponded to which candidate. This undoubtedly caused many voters who intended to vote for Gore to vote for either Buchanan or McReynolds or one of those candidates plus Gore – in which case the ballot would be rejected as an “over-vote”. As noted in “The Miami Herald Report: Democracy Held Hostage”, page 15:


Even the far right wing fringe candidate Patrick Buchanan acknowledged as much, noting that of the 3,407 votes he received in Palm Beach County, only about ten percent of them were meant for him. “The rest, I’m quite sure, were Gore votes”, he said.

Another line of evidence that supports the idea that Gore lost large numbers of votes due to ballot confusion is the fact that thousands of ballots were rejected as over-votes because they contained votes for more than one candidate. This could have happened when a voter felt that s/he had mistakenly voted for one candidate, so attempted to correct the error by voting for the originally intended candidate. With the Palm Beach County butterfly ballot, there were 5,352 over-vote ballots marked for both Gore and Buchanan, in contrast to only 1,676 marked for both Bush and Buchanan (Another 2,864 voters voted for both Gore and McReynolds). All of these ballots were rejected in the final vote count. Making the logical assumption that the vast majority of Gore/Buchanan ballots were meant for Gore and the vast majority of Bush/Buchanan ballots were meant for Bush, Gore would have netted at least 1,500 additional votes in Palm Beach County had these voters voted as intended.

Finally, the most direct evidence of who the voter intended to vote for when an over-vote was produced would be the writing in of a candidate’s name at the bottom of the ballot. Voters wrote in Gore’s name on 2,182 over-votes, while they wrote in Bush’s name on 1,309 over-votes. Many of these ballots involved two votes for Gore – one next to his name and the other for the write-in candidate, where Gore’s name was added by the voter. All of these ballots were excluded from the vote count. Had these ballots on which the intent of the voter was clearly indicated been counted, Gore would have picked up 873 additional net votes, far more than he needed to win the election.

The bottom line of all this discussion with regard to the question of why the election in Florida was initially called for Gore is that many thousands of voters who thought they were voting for Gore actually voted for Buchanan or Reynolds or produced over-votes that were excluded. These voters would have responded to exit polls by saying they were voting for Gore – hence the initial call for Gore. But as the official vote count became more and more complete, the exit polls became irrelevant to predicting the official election results. That is when the call for Gore was reversed. In other words, the “bad data” was the result of voters who intended to vote for Gore and thought they had voted for Gore, but whose vote, whether or not they wrote Gore’s name on the ballot, was never counted for Gore.

Second bad call – due to an electronic “glitch”
The basis of the second bad call, which caused the networks to make the call for Bush as having won both Florida and the national election, at 2:16 a.m. on Wednesday, November 8, can be explained by the following report:


At 2:09 a.m. Volusia County’s erroneous numbers were added to Voter News Service’s tabulations, and less than ten minutes later Florida and the U.S. election were called for Bush. The error in Volusia County had cost Gore (temporarily) 16,021 votes. Another computer error in Brevard County reduced Gore’s vote total by another 4,000 votes.

After the call was made for Bush, as Jeffrey Toobin writes in his book, "Too Close to Call: The Thirty-Six Day Battle to Decide the 2000 election":


The computer glitches in Volusia and Brevard Counties were discovered and corrected, and consequently the TV networks again reversed their call, taking Florida out of the Bush column and calling it “too close to call”, where it remained for several weeks.

The computer error in Volusia Counter was later publicly said to be due to a faulty memory card, with little or no further explanation. It was considered to be accidental, nobody was prosecuted for it, and it stirred up little national attention or concern. The bigger questions are: 1) Were these computer errors purposeful or accidental; and 2) How many other computer errors occurred that night that were not caught?

Probably few people will ever know the answers to these questions. After all, as I will discuss in Chapter 3, the results produced by these electronic voting machines cannot be verified. So unless the errors are massive, as they were in Volusia County, there is little chance of catching them. At least one computer voting expert, Bev Harris, author of Black Box Voting, believes she has good evidence that the error was purposeful, though it didn’t work exactly as planned.


Why did the automatic machine recount reduce Bush’s lead by 80% – from 1784 to 327?

In order to understand the 2000 Presidential election and its importance to future elections in our country, it is instructive to understand the reason why the machine recount reduced Bush’s lead by 80%.

The reduction in Bush’s lead following the machine recount was due almost entirely to the addition of Gore votes in counties that used punch card voting machines. As punch card machines get old they become less efficient at what they are supposed to do – punch holes in ballots. Consequently, use of the older machines often results in failures to fully punch a hole in the card. In some cases the hole may be nearly complete, so that a piece of cardboard (called a chad) is barely hanging from the card, in other cases there may be just a small hole in the card, and in still other cases there may be no tear at all, but only a little dimple. The vote counting component of the machine can only count votes for hole punches that are complete, that is, where any remaining cardboard material is not covering the hole. Thus it is that failures to punch complete holes in the card reduce the vote count. But the act of running cards through the machine a second time tends to cause stray pieces of cardboard that are only loosely attached to the card to fall off, thus enabling the machine to read the ballots as votes where the hole is punched.

For a variety of political reasons, poor areas generally have to make do with older machines than wealthier areas. Consequently, poor areas are generally characterized by a much greater number of undervotes (ballots that register no vote because the hole is not complete) than wealthier areas. Since poor areas tend to vote Democratic, that means that Democratic candidates tend to suffer from this problem. Thus it was that as the cards were run through the machine a second time and many of the hanging chads broke off, the Democratic candidate, Al Gore, gained ground.

This is a problem that almost routinely occurs in almost all elections that use the Votomatic punch card voting machines, not just the 2000 Presidential election in Florida.


In my next post in this series I’ll discuss the highly controversial 36 day Florida recount, which ended when the U.S. Supreme Court abruptly ended the vote count and awarded the Florida and thereby the national election to George W. Bush in what many consider to be one of the three worst U.S. Supreme Court decisions in U.S. history.