Agony of the Arithmecrats

Mr. Dooley was wrong. Politics is beanbag.

Will delegate counting crash the arthmecrats’ computer?

“Just sit back, empty your mind, and let it happen,” I advised political observers on Feb. 6. But did they listen?

The arithmecracy triumphed on Super Tuesday. Hillary Clinton’s alleged momentum was halted, just as Barack Obama’s alleged momentum had been halted in New Hampshire. Even before Super Tuesday, Karl Rove, the designated genius of presidential politics, had written that momentum (he called it “the big bounce”) was a dead letter. After Super Tuesday, everyone agreed that Obama faced a month of likely primary victories—excepting possibly Wisconsin—but that this winning streak would likely end with Clinton winning Texas, Ohio, and Pennsylvania. Nearly all of these predictions have now come to pass. Granted, Obama won Wisconsin by a larger-than-expected 15-percent margin, making significant inroads among white males. But otherwise, this past month has brought no real surprises. Clinton won Texas and Ohio, as she was expected to do, and as I write, polls still indicate she’ll likely win Pennsylvania, too.

But the commentariat has a short memory, even for its own most accurate predictions. To borrow from the joke John McCain likes to tell about his age, political pundits can hide their own Easter eggs. The primary outcomes they foretold came to pass, and lo, they were astounded. They acted like a weatherman who predicts rain on the 6 o’clock news, goes home, goes to bed, wakes up the next morning, looks out the window, and exclaims, “Holy cow! It’s raining!” As Obama piled up one anticipated victory after another, recovering momentucrats fell off the wagon. The Washington Post’s Dan Balz, who on the day after Super Tuesday wrote, “Evenly Matched Dems Portend A Long Race,” a mere one week later wrote, “Clinton Scrambles To Try to Reverse Obama’s Momentum.”

Hillary Clinton encouraged the Obamamentum mirage by replacing her campaign manager, Patti Solis Doyle, right after Super Tuesday. This wasn’t evidence of an imploding campaign so much as what literary critics call an “objective correlative“—an outward, purely symbolic expression of some inward concept or emotion. If you’d skimmed Josh Green’s story about disarray in Clinton’s campaign, posted last month on the Atlantic’s Web site, you might have gotten the impression that Clinton was losing primary after primary to Obama because Doyle had done a poor job. But if you’d read the piece more carefully, you likely would have noticed that, according to Green, Doyle had been doing a poor job for a very long time. Terry McAuliffe, chairman of the Clinton campaign, and Maggie Williams, who eventually replaced Doyle, tried to get rid of Doyle two years ago, only to be overruled by Clinton. This means that if Doyle really did do a poor job (and Green is pretty convincing on this point), she was doing that poor job back when Clinton’s candidacy was riding high. You therefore couldn’t logically blame Obama’s long-predicted victories in places like Louisiana and Virginia and Washington state on “campaign disarray,” except in the negative sense that Doyle failed to turn those states around. (On the other hand, to whatever extent voters find out that Clinton left a purportedly incompetent campaign manager in place two years after she should have canned her, they will likely think less of Clinton’s managerial abilities, which in turn could hurt her in subsequent primaries. That hasn’t happened yet.)

To sum up: The arithmecrats were right all along. Momentum is the fanciful invention of a class of people who have a weakness for (and a financial incentive to create) a dramatic narrative. There is no substitute for counting delegates. Henceforth, we must only count delegates.

But—here’s where it gets complicated—primary delegates alone can’t carry either Obama or Clinton all the way to 2,025, the magic number (a simple majority) needed to win the nomination. It’s mathematically impossible, apparently, given the closeness of the race and the number of primary delegates left to win in the remaining contests. The church of arithmetic therefore will have to yield at some point to the church of perception—in this case, the perceptions of the 795 superdelegates who’ll provide the necessary margin to one candidate or the other.

Superdelegates are free to change their minds about whom to support right up until the roll call at the convention. It’s widely assumed that once the primary season’s over, most superdelegates will side with whichever candidate has won the most popular support. That’s what happened in 1984, according to Walter Mondale’s delegate counter, Tad Devine. Mondale ended the primary season about 40 delegates shy of a majority. Gary Hart had won in more states than Mondale, but Mondale had received slightly more votes overall, and he had more delegates than Hart. So the superdelegates went with Mondale.

The trouble is, how do you define “popular support”? Is it the number of delegates won? According to Slate’s excellent delegate counter, Obama has 140 more primary delegates than Clinton. That spread would narrow to 84 primary delegates if Clinton were to win every remaining primary by 10 points, which seems unlikely. But suppose this were to happen, or some variation that yielded a similar numeric difference. Would superdelegates perceive this as a win for Obama, or a virtual tie? Proportionally, 84 primary delegates would constitute about 2 percent of all delegates in attendance at the convention. Would 2 percent constitute a mandate? How close does the delegate count need to get before the superdelegates call it a wash? A consensus has yet to emerge on that magic number. (Presumably the magic number should be less than the primary-delegate spread between Hart and Mondale in 1984, but I don’t know what that number is. It can’t be 40, because there was a third candidate, Jesse Jackson, who still controlled some delegates. If you know the answer, please e-mail me at

Or is the proper metric the total number of votes cast in the primaries? The Web site RealClearPolitics estimates that Obama is nearly 600,000 primary votes ahead of Clinton nationwide. But if Michigan and Florida are counted (which they probably shouldn’t be, because Obama and Clinton didn’t campaign there and because in Michigan Hillary was the only major candidate on the ballot), that 600,000 plurality shifts over to Clinton. Conceivably the same percentages might be achieved if Michigan and Florida are permitted to vote again prior to the convention, as seems more likely. It’s even possible that one candidate will receive a plurality of primary votes while another candidate receives a plurality of delegates. Do votes trump delegates? Do delegates trump votes?

A cherished political cliché has it that politics ain’t beanbag. But Mr. Dooley was wrong. Politics is beanbag, not in the sense that it’s a gentle game suitable for small children but in the sense that you don’t necessarily have to get the beanbag through the hole in the target box to win. Under the right circumstances, you can still win if you get it near the hole. To pull that off, though, it helps to know the rules. The arithmecracy has no clear guidance to offer, because these aren’t mathematical questions.

Momentucracy vs. Arithmecracy Archive:

Feb. 6, 2008: “Triumph of the Arithmecrats

Feb. 1, 2008: “On the Media” interview about momentucracy and arithmecracy, New York Public Radio

Jan. 30, 2008: “Momentucrats vs. Arithmecrats, Part 2

Jan. 28, 2008: “Momentucrats vs. Arithmecrats

Jan. 21, 2008: “Is Obama Winning?”

Dec. 11, 2007: “Whose Nominee Is It, Anyway?”