In all likelihood, George W. Bush still would have won Florida and the presidency last year if either of two limited recounts -- one requested by Al Gore, the other ordered by the Florida Supreme Court -- had been completed, according to a study commissioned by The Washington Post and other news organizations.
But if Gore had found a way to trigger a statewide recount of all disputed ballots, or if the courts had required it, the result likely would have been different. An examination of uncounted ballots throughout Florida found enough where voter intent was clear to give Gore the narrowest of margins.
The study showed that if the two limited recounts had not been short-circuited -- the first by Florida county and state election officials and the second by the U.S. Supreme Court -- Bush would have held his lead over Gore, with margins ranging from 225 to 493 votes, depending on the standard. But the study also found that whether dimples are counted or a more restrictive standard is used, a statewide tally favored Gore by 60 to 171 votes.
Gore's narrow margin in the statewide count was the result of a windfall in overvotes. Those ballots -- on which a voter may have marked a candidate's name and also written it in -- were rejected by machines as a double vote on Election Day and most also would not have been included in either of the limited recounts.
The study by The Post and other media groups, an unprecedented effort that involved examining 175,010 ballots in 67 counties, underscores what began to be apparent as soon as the polls closed in the nation's third most populous state Nov. 7, 2000: that no one can say with certainty who actually won Florida. Under every scenario used in the study, the winning margin remains less than 500 votes out of almost 6 million cast.
For 36 days after the election, the results in Florida remained in doubt, and so did the winner of the presidency. Bush emerged victorious when the U.S. Supreme Court, in a 5 to 4 ruling, agreed with his lawyers' contention that the counting should end. Since then, many Gore partisans have accused the court of unfairly aborting a process that would have put their candidate ahead.
But an examination of the disputed ballots suggests that in hindsight the battalions of lawyers and election experts who descended on Florida pursued strategies that ended up working against the interests of their candidates.
The study indicates, for example, that Bush had less to fear from the recounts underway than he thought. Under any standard used to judge the ballots in the four counties where Gore lawyers had sought a recount -- Palm Beach, Broward, Miami-Dade and Volusia -- Bush still ended up with more votes than Gore, according to the study. Bush also would have had more votes if the limited statewide recount ordered by the Florida Supreme Court and then stopped by the U.S. Supreme Court had been carried through.
Had Bush not been party to short-circuiting those recounts, he might have escaped criticism that his victory hinged on legal maneuvering rather than on counting the votes.
In Gore's case, the decision to ask for recounts in four counties rather than seek a statewide recount ultimately had far greater impact. But in the chaos of the early days of the recount battle, when Gore needed additional votes as quickly as possible and recounts in the four heavily Democratic counties offered him that possibility, that was not so obvious.
Nor was there any guarantee that Gore could have succeeded in getting a statewide recount. Florida law provided no mechanism to ask for a statewide recount, only county-by-county recounts. And although he did at one point call on Bush to join him in asking for a statewide recount, it was with the condition that Bush renounce all further legal action. Bush dismissed the offer, calling it a public relations gesture by his opponent, and Gore never took any further steps toward that goal.
White House press secretary Ari Fleischer, responding to the study, said, "The voters settled this election last fall, and the nation moved on a long time ago. The White House isn't focused on this; the voters aren't focused on it." Fleischer called the results "superfluous."
Gore, in a written statement, did not respond directly to the study. "As I said on Dec. 13th of last year, we are a nation of laws and the presidential election of 2000 is over," he said. "And of course, right now our country faces a great challenge as we seek to successfully combat terrorism. I fully support President Bush's efforts to achieve that goal."
Gore said he remained appreciative of the support he received last year and "proud of the values and ideals for which we fought."
Discerning Voter Intent Conducted by the National Opinion Research Center, an organization based at the University of Chicago, the study examined all ballots that were initially rejected by voting machines. This included those that contained no discernible vote for president, known as "undervotes," and those that registered votes for more than one candidate, the "overvotes."
Last year's recount battles largely focused on about 61,000 undervote ballots. In the recounts, Gore advisers pushed for the most liberal interpretation of voter intent, giving rise to heated disputes and legal wrangling over whether "dimpled chads" on punch-card ballots should be counted as votes.
But in another twist clear only now, the study found that where Gore had the greatest opportunity to pick up votes was not in those undervote ballots but in the approximately 114,000 overvote ballots, particularly 25,000 overvote ballots read by optical scanning machines.
Using the most inclusive standards, Bush actually gained more votes than Gore -- about 300 net -- from the examination of the undervote ballots. But Gore picked up 885 more votes than Bush from the examination of overvote ballots, 662 of those from optical scan ballots.
The study did not credit Gore with the thousands of votes lost as a result of the infamous butterfly ballot in Palm Beach County. Many voters using the ballot became confused by the listing of presidential candidates on two facing pages and punched Gore's name and one of the candidates next to him, nullifying their vote.
An examination of the Senate choices on those ballots indicates the mistakes were made overwhelmingly by Democrats and suggests that Gore lost about 8,000 votes because of the confusion. The Post study did not award those overvotes to Gore because no clear voter intent could be determined on a ballot where two candidates were marked. A similar analysis of the two-page presidential ballot in Duval County showed Gore lost about 7,000 votes, which also could not be given to Gore in the study.
Gore never pushed hard for the kind of full recount that might have brought overvotes into play. And the Florida Supreme Court, which on Dec. 8 ordered a statewide manual recount -- halted in midstream the next day by the U.S. Supreme Court -- focused on undervotes and required only that undervotes be retabulated.
Ironically, it was Bush's lawyers who argued that recounting only the undervotes violated the constitutional guarantee to equal protection. And the U.S. Supreme Court, in its Dec. 12 ruling that ended the dispute, also questioned whether the Florida court should have limited a statewide recount only to undervotes.
Had the high court acted on that, and had there been enough time left for the Florida Supreme Court to require yet another statewide recount, Gore's chances would have been dramatically improved. But there are too many variables in any effort to reexamine the ballots -- from varying standards in judging ballots in the counties to problems of getting an exact replication of the overvote and undervote ballots -- to be able to say with absolute certainty what might have happened in Florida.
"In my opinion, it's too close to call," said Kirk Wolter, senior vice president of NORC. "If we take it as given that two major candidates were separated by perhaps a few hundred or fewer ballots, it may be that we'll never know the exact vote total."
Historical Record Designed to provide a historical record for one of the most remarkable presidential elections in U.S. history, the ballot study was launched early this year by a consortium of news organizations and originally was to have been completed by last spring. Consortium members, in addition to The Post, included the New York Times, the Associated Press, CNN, the Los Angeles Times, the Wall Street Journal and four Florida newspapers: the Orlando Sentinel, the Palm Beach Post, the South Florida Sun-Sentinel in Fort Lauderdale and the St. Petersburg Times.
"We joined the consortium to obtain an accurate, nonpartisan assessment of the uncounted ballots in Florida to determine how the people of Florida voted and why their voting systems did not work better," said Post Executive Editor Leonard Downie Jr. "The results shed light on the actions of the players in the constitutional drama in Florida. They also provide information that can help the federal and state governments improve voting systems nationwide. And they will help historians better analyze a unique and important event in American history."
Various technical problems delayed the study, including the difficulty county officials had in separating the disputed ballots into undervotes and overvotes. The events of Sept. 11 set back publication further because news organizations were devoting all their resources to coverage of the terrorist attacks and subsequent events.
The project used impartial observers hired by NORC to examine the ballots and considered many possible alternatives for tallying the votes. But no study of this type can accurately recreate Election Day 2000 or predict what might have emerged from individual battles over more than 6 million votes in Florida's 67 counties.
Three individuals, operating independently, examined each undervote ballot and some of the overvote ballots. However, most of the overvote ballots, which are less subject to different interpretation over their markings, were viewed by one person. The Post's findings are based primarily on results in which two of the three reviewers agreed on the marks on the ballot, deemed a fair standard for discerning what was on the ballot.
The new study differs from an earlier ballot examination by the Miami Herald and USA Today, which did not systematically look at all overvote ballots, instead relying on a computer analysis of those ballots. In that study, one person, usually an accountant, determined marks on individual undervote ballots. A second person also looked at the undervote ballots, but the accountant's coding was always used if they differed. The study concluded that Bush would have won under almost all situations.
The NORC observer teams hired by the consortium did not decide whether the undervote or overvote ballots would have been counted as valid votes in a recount. Instead, they worked independently, using a coding scheme to describe the marks on each ballot under supervision of a NORC team leader.
The study projects possible election outcomes based on different scenarios -- which ballots might have been included in recounts and what marks on those ballots might have been considered as votes.
On ballots from punch-card machines, such as those used in the South Florida counties where Gore asked for recounts, these marks included a dimpled chad, which is the appearance of an indentation, or chad with one or more corners detached.
On ballots from optical scanning machines, the marks included instances where a voter circled or wrote in the candidate's name rather than filling in an oval next to the name on the ballot.
The Post, in conjunction with the other news organizations, reviewed the descriptive codes to apply different standards for determining voter intent and tallied results based on several scenarios that sought to approximate conditions on the ground in Florida.
The three examiners agreed most of the time, but Post analysis of ballot swings caused by disagreement showed more than enough votes to decide the election.
The Winner Bush was certified by the Florida election canvassing commission as the winner by 537 votes Nov. 26. That certification came after Gore had asked for recounts in Volusia, Broward, Miami-Dade and Palm Beach counties. But it included full results from only Volusia and Broward, which met the state's 5 p.m. deadline.
Palm Beach County submitted its final results about two hours past the deadline, but Florida Secretary of State Katherine Harris declined to include them. Officials in Miami-Dade halted their recount days earlier, amid GOP-inspired protests, claiming they would not have enough time to meet the state's deadline.
Had all four counties completed their recounts, as requested by Gore, and been included in the state certification, Bush still would have been declared the winner, but by just 225 votes, according to the analysis by The Post and other news organizations.
The Florida Supreme Court's Dec. 8 order for a statewide manual recount of all undervote ballots also would have resulted in Bush as the winner, the study found. Gore's team protested when the U.S. Supreme Court on Dec. 9 agreed to the Bush campaign's request for a stay, halting that recount in midstream. But the study found that a count of all undervotes in the state would have left Bush ahead of Gore by 430 votes.
Some counties ignored the state Supreme Court order that weekend and refused to conduct manual recounts. Other counties included undervote and overvote ballots in their recounts. The media consortium surveyed the counties to determine what standards they were using. On the basis of those standards -- the closest approximation possible to what was happening that weekend -- the Post study found that, if the court had not intervened to stop the counting, Bush would have won by 493 votes.
But the results in Florida and, therefore, in the presidential election might have been different had the 67 counties been ordered to proceed with a manual recount of all undervotes and overvotes.
Under several scenarios examined by the consortium, and using a standard in which two of the three reviewers agreed on the markings on each ballot, Gore emerged with more votes than Bush.
The overvotes that could have provided the margin for Gore were on ballots where voters tried to be extra-clear in their choice and ended up nullifying the vote. They filled in the oval next to a candidate and then filled in the oval for "write-in" and wrote the same candidate's name again.
Those overvotes were rejected by machines, but some county officials examined those ballots on election night to reclaim the votes. Other counties, though, didn't check for those obvious votes. Gore had more than 500 of those votes in Lake County and more than 250 in Escambia, netting him gains of 172 and 157 votes against Bush in those counties.
The narrowest margin, according to the study, came under a scenario in which at least one corner of a chad was detached from punch-card ballots -- the prevailing standard across the state of Florida at the time -- or any mark on the optical scan ballots showing clear voter intent. In that case, the study showed Gore with 60 votes more than Bush.
Gore's margin grows under three other scenarios. Under the least-restrictive standard for interpreting voter intent, which counted all dimpled chads and any discernible optical mark (which in the case of optical ballots Florida's new election law now requires to be counted as votes), Gore had 107 more votes.
Gore's margin rose to 115 votes in the study under a tighter standard, calling for chads to be fully punched and a more restrictive interpretation of what constitutes a valid mark on optical scan ballots.
But this is one case where disagreements among the reviewers affected the outcome. Gore won under this scenario when two of the reviewers agree on the markings. Under a standard in which all three were required to agree, Bush won by 219 votes.
Gore's largest margin in a statewide recount involving all ballots comes under a scenario that sought to recreate the standards established by each of the counties in their recounts. In that case, Gore emerged with 171 more votes than Bush.