Is the election over?

I'm no conspiracy theorist, and I like to think I'm a realist. And as I've mentioned before, Kos has a few too many left-wing wackos posting far-out diaries. But one intriguing possibility raised over and over again on liberal boards simply is not dying: that somehow, Kerry can still win Ohio's 20 electoral votes. Most amazingly, these claims have a sliver of validity. There's too much conflicting information floating out there to make any definitive claims about an election reversal in the Buckeye State, but we do know a few things. One: the 'real' presidential election hasn't happened yet. That happens on Dec. 13, when the slate of electors for Ohio and the other 49 states meets in D.C. to 'officially' cast their votes for president. And those votes, in turn, are not certified until Congress opens them on January 6. Though the Ohio Secretary of State plans to have the election finished on December 3, that is clearly not a drop-dead date in the constitutional sense. Two: In the initial count, Bush won Ohio by a vote of 2,796,147 to John Kerry's 2,659,664 -- a margin of about 136,000. Three: right now, with Ohio, Bush would win with 286. Kerry has 252. If Ohio flips, Kerry gets 272 EVs (above the 270 threshold), and the presidency. Four: The day after the election, Kerry conceded Ohio and the election. Five: The Green and Libertarian Parties have ponied up enough money in Ohio to start the recount process. Given those facts, the relevant number appears to be 136,000. What are the chances that margin, far greater than the number that delivered Florida for Bush in 2000, can be eliminated? Not very good, but here is the formula. There are 155,000 'provisional' ballots to be counted. Without getting into the vagaries of Ohio election law, these ballots were provided to voters who were not found on the voting rolls of their home precincts when they went to vote; they will count if it is discovered they were genuinely allowed to vote in that precinct (I'm not getting into the idiocy of denying the vote because the voter is at the wrong polling station, but there you are.) It was believed that many of these provisional ballots would be thrown out, but it appears as though they are being admitted as valid at about the same rate they were in 2000. According to one Ohio paper, "In the 11 Ohio counties that have finished checking provisional ballots cast in the presidential election, 81 percent have turned out to be valid. It is too early to know whether the ballots have benefited Bush or Kerry because counties first need to determine their validity before conducting the count." Suppose the statewide stack of provisional votes were valid at the same rate as in those 11 precincts -- that would mean 125,550 more votes, either way, into the mix. The Internet consensus is that 85 percent of provisional ballots went Gore's way in 2000. I haven't seen any hard citation of that number, so I'll have to assume it's high. It does pass the muster of common sense that Kerry would win the provisional ballots though: they do tend to be cast in urban rather than rural districts because urbanites tend to change addresses more often. Let's say they go 70-30 Kerry, a conservative estimate. 87,885 more votes for Kerry; 37,665 Bush; the margin for Bush drops to 75,330. Here's where things get interesting. First, preliminary examinations of voting machine errors show that the vote totals are already changing. From the same Ohio paper: "According to a November 5 article by the Associated Press, elections officials admitted that an error with an electronic voting system gave President Bush 3,893 extra votes in a Gahanna precinct. Franklin County reported Bush with 4,258 votes and John Kerry with 260, even though only 638 voters cast ballots in that precinct. Election officials in that county now say a cartridge from a voting machine generated errors after the precinct closed, and only 365 people voted for Bush." That's 7,786 votes gone from Bush's column right there, bringing the margin down to 67,544. Second, there were, apparently, 93,000 'undervotes' in Ohio -- that is, ballots cast without any presidential vote counted. If you remember Florida 2000, you'll recall that this phenomenon occurs with the dreaded punch card ballots when the hole is not completely punched, either because too many loose chads remain underneath the hold or through voter error. The vast majority of these punch card machines were in strong Democratic-leaning urban areas. I'm not even going to venture a guess as to whether these go for Kerry, let alone by how much. The undervotes will be counted pretty liberally (no pun intended) -- apparently Ohio election law will count such a ballot if there's even a trace of a vote. Of course, these only get counted in a recount. Given that it's unlikely many people abstained from voting for president when they cast a ballot, if a high proportion of those ballots turn out to be valid, and a large majority vote for Kerry (again, both unverifiable propositions at this point), then we could be headed for a Florida 2K-style circus. These numbers could be fudged up or down either way, but I think it's pretty safe to say that Bush's margin of victory in Ohio will not be as high as 130,000 when all is said and done. I'd give Kerry a one-in-a-million shot of getting to the White House this time around. But as Lloyd so famously gushed over similar odds in Dumb and Dumber, "So you're telling me there's a chance!"