Nearly 1.75-million Democrats voted in Florida's officially meaningless presidential primary, and it remains to be seen whether Florida will send any delegates to the national convention.
But in this Twilight Zone of a campaign, a more cosmic question arises: Do the Florida votes actually exist?
In Barack Obama world the answer is no. Clinton's 295,000-vote victory margin in Florida is basically imaginary. The candidates chose not to campaign in the state, the argument goes, so the votes should be purged from the minds of any uncommitted superdelegate.
On planet Hillary Clinton, those Florida votes are as real as the White House itself. We cannot pretend a record turnout of voters never happened.
This Florida riddle gained importance when Clinton beat Obama on Tuesday in Pennsylvania, another big swing state. Obama's lead over Clinton in pledged delegates may be insurmountable, but Clinton might still overtake him in the popular vote count, which doesn't really matter except that it would be a persuasive argument for Clinton to make with the "superdelegate" party officials who will choose the nominee.
But to catch Obama's vote total, Clinton needs Florida's ballots.
"Those votes count,'' the New York senator insisted to the St. Petersburg Times recently. "They're part of the popular vote. You can't erase them. They were certified. The secretary of state said they happened; 1.75-million people showed up. I believe that they were part of this primary election season."
In an already confounding election, however, measuring the popular vote can be a murky exercise. Consider:
• Michigan, like Florida, lost its delegates for violating the national party rules by holding a January primary, and Obama actually took his name off the ballot to avoid antagonizing party leaders in early election states such as Iowa and New Hampshire. The result was Clinton 55 percent and "uncommitted" 40 percent.
• Four states — Iowa, Nevada, Maine and Washington — held caucuses where the overall vote totals never were reported.
• In Florida, no delegates were at stake, and the candidates did not campaign.
The respected Web site RealClearPolitics.com finds the clearest cut popular vote tally gives Obama a lead of 500,353 votes. Extrapolating totals for Iowa, Nevada, Maine and Washington, Obama's lead rises to 610,575.
Throw in Florida, and Obama's lead drops back down to 315,803. Then add in Michigan, giving Clinton the votes for her and Obama zero votes, and suddenly Clinton takes the popular vote lead by 12,506 votes, or .04 percent.
If she can maintain that lead through the nine contests between now and June 3 she can try to argue to superdelegates that not only is she winning more crucial, swing states than Obama but that she has won more votes overall. Of course, that's no certainty, as Obama is heavily favored in the May 6 primary in North Carolina, the most populous state remaining.
Candidates win the nomination with delegates, not popular votes. That's why the Obama campaign shrewdly targeted its resources toward specific congressional districts and low-turnout caucus elections to maximize his delegate haul in state after state.
"If this was about popular vote ... we would have gone and lived in California and tried to run up the score in Illinois and spent more time in New York and New Jersey,'' Obama campaign manager David Plouffe said Wednesday.
"But we constructed our strategy based on pledged delegates, as did she.''
But neither Obama nor Clinton can win without winning over undecided superdelegates, who are free to consider any subjective measurement they want — including the disputed votes in Florida and Michigan.
"If the gap between Clinton and Obama is 300,000 votes and those Florida votes could have made the difference," said U.S. Rep. Jason Altmire, a Democrat from western Pennsylvania whose district easily chose Clinton, "then some people are going to use that."
The Obama campaign is pushing back hard on any suggestion that Florida or Michigan should count for anything.
Michigan is a different matter, but in Florida, Obama and Clinton were both on the ballot, and a record number of Democrats turned out with strong encouragement from the Florida Democratic Party.
Regardless of delegates awarded, many Democrats dispute the notion that the election was somehow an invalid reflection of voter sentiment because the candidates did not stage rallies or pay for robo calls.
"Was somebody living in a cave? Didn't they turn on Good Morning America and the Today Show?'' asked Sen. Bill Nelson, a Clinton supporter who said superdelegates "absolutely" should consider the results in Florida.
"If the people in Florida either read a newspaper or turned on a television and/or a radio ... they certainly were subjected to the campaigning that was occurring and indeed responded."
Muriel K. Offerman, a superdelegate and DNC member from Raleigh, N.C., said that she may consider the outcomes in Florida and Michigan when deciding whom to support but that she'll keep in mind that Obama "absolutely" would have done better against Clinton had he campaigned there.
"It's hard to ignore them, because so many people voted. But again, you hope you can ignore them on the one hand – it was against the rules of the party," she said. "Whether I consider half the numbers, or two thirds, I don't know yet."
Times staff writer Wes Allison contributed to this report. Adam C. Smith can be reached at email@example.com or (727)893-8241.
Correction: A previous version of this story indicated Hillary Rodham Clinton would claim the lead in the national popular vote if Michigan results were counted and the nearly 240,000 "uncommitted" votes were given to Barack Obama. But for Clinton to claim the lead, Obama would have to be given zero votes from Michigan, where his name was not on the ballot.