The Washington Post has discovered, to its horror, that despite all the election “reforms” since 2000, there is a good chance “the unthinkable” will happen again in 2004: “Americans may not know who won the presidential race on Tuesday night. Again.”

There are several problems with this analysis. First of all, the recent election reforms have nothing whatsoever to do with assuring a speedy result — only an accurate one. And in this context, speed and accuracy are often competing factors rather than complementary ones; take, for example, provisional ballots, mandated by the Help America Vote Act, which (hypothetically) increase accuracy but (most assuredly) decrease speed. So the mention of the HAVA reforms in the opening paragraph is a total non-sequitur.

Secondly, “not know[ing] who won the presidential race on Tuesday night” is by no means unthinkable. In fact, it’s not even horrifying. Not knowing who won the presidential race on Jan. 20 — now that would be horrifying. But even that scenario isn’t “unthinkable” (as evidenced by the fact that I just thought of it). The prospect of not knowing the winner Tuesday night is, at most, mildly disconcerting.

My third objection to the Post article is along a somewhat different line. I don’t think the “we won’t know the winner for a long time” scenario is actually as probable as many pundits seem to think.

Listening to the hype, you’d think it’s almost inevitable, but I just don’t believe that’s so. Whereas the (false) assumption in the past was that we’d always know the result of every state’s vote within a few hours of the polls closing, the (false) assumption now has lurched to the opposite extreme: the result in any state that is remotely close will be unknowable for days, weeks, or months. That’s simply not true.

Look at the 2000 election timeline. By the time the nation woke up Wednesday morning, there were only three states whose outcomes were not known: Florida, New Mexico and Oregon. Florida and New Mexico were unknown because the margins of victory there were less than six hundred votes. Oregon was unknown because of lots and lots absentee ballots.

All the other close states, however, were “called” — accurately — by 6:22 AM EST at the latest. Michigan and Pennsylvania, supposedly the biggest “battlegrounds” aside from Florida, were both called within an hour of the polls closing (based primarily or exclusively on exit polls). New Hampshire, which Bush won by 1.27%, was called at 10:05 PM. Iowa, which Gore won by 0.31%, was called at 2:04 AM. Wisconsin, which Gore won by 0.22%, was called at 6:22 AM. All of those calls were correct, and all of them took place before the nation arose from its slumber Wednesday morning.

The only reason the entire election remained undecided for more than a month was because a) the result hinged on a single state, and b) that state was decided by an unbelieveably small margin. That scenario could repeat itself this year, but it is far from inevitable. It isn’t enough that the election is “very close.” The stars really have to align in a unique way to reproduce what happened in 2000. People are losing sight of that in all this pre-election hype.

Imagine, for example, that the three undecided-after-7am states in 2000 had been New Mexico, Oregon and Iowa, instead of New Mexico, Oregon and Florida. Had that been so, the election would not have dragged on the way it did. The fact that three states were up in the air wouldn’t have mattered, because it wouldn’t have affected the electoral-college math in a way that changes the result. The winner in Florida would have won the presidency, regardless of those other states. So again, it isn’t enough that the electoral vote be close; the numbers have to add up just so. That’s possible, but by no means inevitable.

Now, granted, the litigation this time around is likely to be much more widespread than in 2000; lawsuits will probably be filed even in states where the margin is considerably less close than it was in Florida. (More on that problematic issue in a forthcoming post.) But the fact is, recounts and lawsuits are very unlikely to change the outcome of a state where the margin is more than a few tenths of a percent. (Florida’s margin, remember, was less than one one-hundredth of a percent, and even there the recounts and lawsuits didn’t, in the end, change the outcome.) More likely, lawsuits and recounts in such a situation will just prolong the inevitable, giving partisans on the losing side an opportunity to bitch and moan, but to no actual effect.

So perhaps the most likely the result is that we’ll have a “probable winner” for a few days or a couple of weeks, rather than a definite, 100% certain winner. But I doubt we’ll have a total deadlock for more than a month, a seemingly endless no-idea-who-won situation, like we did in 2000. It could happen, but it’s not nearly as likely as people think.