October 31, 2004

Will We Know Who the Next President is by Wednesday?

With just about all of the national polls showing the race dead even or within the margin of error, it is tempting to overestimate the chances of post-election litigation. I will continue to stick by my prediction that there is still only a 10% chance that we see a delay in knowing the winner of the election because of post-election litigation.

As I've said, for there to be post-election litigation that is outcome determinative, we will need a sufficiently large problem in a battleground state whose electoral votes will matter for the electoral college tally. How likely is that?

First, it is not all that clear that the election will be very close in absolute terms. True, the polls show a dead even race. But look at the figures from 2000. A Gallup tracking poll from Oct. 29, 2000 had Bush over Gore by 7. With all the new registrations this year, the increased use of cell phones, lots of people screening phone calls, and many anecdotal stories about increased turnout, polling based on likely voters may well be off by at least a few percent.

Historically, as Lawrence Jacobs has noted, "only two of the last 10 presidential elections had total voting margins of less than 1 percent...and even in those races the differences in close states were measured in thousands or tens of thousands of votes." We would have to have another razor-thin race in absolute terms, as last time, where, according to this chart, two states were within 1,000 votes and two more states within 6,000 votes.

Even if one state is that close, it would have to be a state that matters to the outcome of the Electoral College. Many of the Electoral College predictions have also been razor thin, but the predictions are volatile. The Slate daily prediction has vascillated between slightly Bush and slightly Kerry all week now shows Kerry with a 299-239 lead (though noting "his lead is shakier than it looks"). It is possible that either candidate ends up with an electoral college landslide if the popular vote polls are off by just a percentage in one direction or another.

The most likely candidates for a state that could plunge the country into litigation chaos are battleground states with the most votes and that remain very close, mostly Ohio and Florida. In September I called Ohio the most likely candidate for post-election problems and I still believe that. But the chances of problems in Ohio have been somewhat minimized by pre-election litigation. Remember, that courts have resolved issues in Ohio related to wrong precinct votes, voter identification, pre-election day challenges of voters, and other issues. By tomorrow, we expect rulings on how election day challenges will be conducted there.

Of course, if the margin in Ohio is razor-thin in absolute vote terms (i.e., votes within the thousands or fewer), and Ohio's votes are determinative of the outcome of the election, post-election litigation is all but guaranteed. There can be mini-wars fought over many of the provisional ballots, and they will be fought. Along the way, there will be charges of equal protection violations (are elections officials throughout Ohio using the same standards for deciding which provisional ballots should be counted?) and allegations of fraud and vote suppression. These are all fact intensive questions that could gum up the works for weeks.

So we are not out of the woods by any means. Still, it would take another "perfect storm" to make the courts, rather than the people, in the position of ultimate arbiters.

In the short term, there's not much we can do now. In the longer term, we need some radical surgery for our system of election administration. More about that to come.

Posted by Rick Hasen at October 31, 2004 10:41 PM