As Kos detailed yesterday, even with the superdelegates going as their states go, Obama wins. The best possible fallback argument for Clinton: the superdelegates should support the candidate that received the largest national popular vote total. Yet she's trailing in the popular vote any way you slice it. How to turn the tide?
According to Real Clear Politics, Obama leads the popular vote (excluding Florida and Michigan) 13,000,655 to 12,411,705 or by 588,950.
The solution for Senator Clinton: a mail-in vote in Florida, where turnout numbers are likely to be much, much higher.
According to Clinton campaign insider Bill Nelson, who appears to be running the show in Florida:
The details have yet to be worked out precisely, according to Sen. Bill Nelson (D-FL), but it would involve a mail-in vote administered by the state party. "My job is clear," Nelson told Newsweek. "It's to stand up for the right of Floridians to vote as intended."
Clearly the best case scenario for Senator Clinton is a mail-in vote in Florida, where she can run up the score in the popular vote totals, while agreeing to a caucus in Michigan, which will result in the smallest number of participants in a state she very well may lose.
It is difficult to predict the extent to which vote-by-mail will increase the turnout in Florida. One study has pegged it at 10%.
In elections that capture the public's imagination, of which this is undoubtedly one, history suggests it could be much greater. The only historical reference I can think of is the 1996 special election in Oregon between Gordon Smith and Ron Wyden for Bob Packwood's empty Senate seat. In that highly publicized race, and the first Senate race conducted by mail, two-thirds of Oregon's voters cast their ballots by mail, up from the 38% who voted in 1994 and from the 50% that voted in the 1992 presidential election.
So where are we? Let's assume that Senator Clinton wins a vote-by-mail primary in Florida by the same margin she won the first time, 60.18% to 39.82%. The first time around, 1,447,200 voted in the Florida primary, with Senator Clinton taking 870,986 popular votes to Senator Obama's 576,214, for a difference of 294,772. But that still finds her trailing by 294,178 (excluding Michigan). So if we redo Florida, Senator Clinton needs to gain 588,950 votes to close the gap.
Forgetting the upcoming contests for the moment, how many more people have to show up for her to close the gap?
100% More
If the margins remain the same and twice as many people vote than the first time around, Senator Clinton's total votes go from 870,986 to 1,741,972, while Senator Obama's total votes go from 576,214 to 1,152,428, for a difference of 588,950 votes.
What is the likelihood that twice as many Floridians vote this time around AND Senator Obama doesn't gain a higher percentage of Florida's support than on January 29? I'll leave it to others to decide. While not the topic of this diary, keep this in mind as well:
On a state-by-state basis, Democrats had higher turnouts than Republicans in 19 out of 25 states. The six outliers that tilted Republican were Arizona (Sen. John McCain's home turf), Utah (a pro-Romney Mormon stronghold), Michigan (where Rep. Denis Kucinich was the sole Democrat to campaign), Florida (where no Democrats campaigned), Alabama, and Alaska.
If one accepts that a truly contested primary--with all the campaigning that goes with it--would result in greater turnout and that a mail-in vote would juice turnout even more, twice the number of votes is not inconceivable.