April 4, 2001 -- George W. Bush still would have gotten more votes than Al Gore even if the U.S. Supreme Court hadn't halted the manual recount in Florida, according to a comprehensive analysis of uncounted ballots.
In fact, Bush's 537-vote margin of victory would have increased to 1,665 under the ballot-counting standards Gore's supporters had advocated, according to a review conducted by The Miami Herald, its parent company Knight Ridder, and USA Today.
The analysis, which looked at 64,248 uncounted ballots in all 67 of Florida's counties, has Republican Bush winning under most scenarios.
"This is very good news for Bush," ABCNEWS political analyst George Stephanopoulos said on Good Morning America. "In almost all the scenarios, Bush wins."
Stephanopoulos said that the study undercuts Democratic arguments that Republicans stole the election.
A Little Bit of Irony
"What we do know is that the process that was going forward in December that was stopped by the U.S. Supreme Court would have resulted in a Bush victory, said Miami Herald Managing Editor Mark Seibel. "Bush, under most scenarios, would have won a recount of the Florida ballots as the Florida Supreme Court ordered."
But the analysis also found that Gore would have won by three votes if the recount had continued under the standards advocated by Bush supporters.
Seibel noted the irony in that fact.
"The irony, of course, is that the Bush people were always saying you can't count those things," said Seibel. "And the Gore people insisted that they would. But in these counties, it particularly benefited Bush."
Hard to Say What Would Have Happened
While the analysis shows that Bush would have won Florida in all but a few cases, it is hard to say whether that would have happened in reality if the hand counting had continued.
The analysis looked at multiple scenarios for a recount, but in each case applied the same standards to all the ballots. In reality, each county would have been allowed to determine its own standards. Some counties actually sought clarification from the courts on just what standards should be applied to the ballots, but did not get it.
And what will never be known is what the different county canvassing boards — and individual board members themselves — would have found looking at the same ballots.
Four Scenarios, Three Bush Wins... and a Question Mark
The analysis, conducted by public accounting firm BDO Seidman, LLP, projected the vote total based on several different recount standards.
During the monthlong post-election recount drama, Gore supporters argued for a widely inclusive standard, under which ballot-counters were asked to divine the intent of the voted based on "hanging" chads, "pregnant" chads and "dimpled" chads.
Chads are the little pieces of paper that are supposed to be punched out of the ballot after the vote is cast.
It is under those most generous, Gore-advocated, standards that the Bush margin of victory rises to 1,665.
Bush supporters argued that the recount should have been stopped altogether, but said that if it did continue, only ballots in which the intent of the voter is clear — ie., a chad-free ballot with only one presidential candidate selected — should be counted.
Under that scenario, Gore would have won by three votes, according to the analysis. But, the study cautions, that would have given the former vice president a 0.00005 percent margin of victory, which could have raised even more questions.
If dimpled presidential votes were counted only in instances where votes in other races on the same ballot were also dimpled — suggesting a problem with the machinery — Bush would have won by 884 votes.
Under a scenario in which chads had to be detached by at least two corners for the vote to count, Bush's margin of victory decreases to 363.
Some Good News for Gore
While the analysis showed the Bush would have won if the court-ordered recounts under way at the time of the U.S. Supreme Court ruling had been allowed to continue, it was not a clear win for the GOP.
The study also looked at four scenarios in which the entire Florida undervote was counted — something that was not ordered by any court — and found that Gore would have won under two of those.
This latest analysis will not be the last word on the Florida recount: the National Opinion Research Center at the University of Chicago is conducting a separate vote-counting project on behalf of a number of media outlets.
ABCNEWS political director Mark Halperin, polling unit director Gary Langer and ABCNEWS Radio contributed to this report.