There's a new candidate in the Democratic race. The candidate's name: "expected." Winning isn't enough. You have to do better than "expected."

"Expected" often plays an important role. Take the case of the 1992 New Hampshire primary. After the Gennifer Flowers and draft record controversies, Bill Clinton was in grave danger in the Granite State. He ended up finishing second to Paul Tsongas. But that placement was better than "expected." So he was able to proclaim himself "the Comeback Kid."

Polls help set expectations, even though they're not particularly good at predicting the results of primaries and caucuses—multicandidate races where turnout is volatile. Right now, polls tell us to expect a close race between Dick Gephardt and Howard Dean in the January 19 Iowa caucuses. If either wins big, he will have done better than expected.

Dean appeared to be losing some momentum in Iowa last week, as news got out that he had once criticized the Iowa caucuses as being "dominated by special interests." Then on January 9, Dean won the endorsement of Iowa's most popular Democrat, Sen. Tom Harkin. The former Vermont governor was expected to regain momentum, at least in Iowa.

Iowa is critical for Gephardt. He comes from neighboring Missouri and won the Iowa caucuses in 1988. If Gephardt doesn't win Iowa this time, he's expected to have trouble staying in the race. With Wesley Clark and Joe Lieberman opting out of the Iowa contest, a tight race is shaping up for third place between John Kerry and John Edwards. A second-place finish by either of them would be an unexpectedly strong showing.

Meanwhile, in New Hampshire, polls are setting high expectations for Dean's performance in that state's January 27 primary. If Dean doesn't win New Hampshire by a big margin, he'll do worse than expected. At one time, Kerry, who is from neighboring Massachusetts, was expected to win New Hampshire handily. But Kerry has sunk so low in the polls that if he comes anywhere close to Dean, he'll have done better than expected.

Meanwhile, Clark, Lieberman, and Edwards are all hoping to do unexpectedly well in New Hampshire, and maybe even come in second to Dean. If Clark finishes ahead of Kerry, Kerry's campaign could be finished.

After New Hampshire, the real contest begins: the one between Dean and "Stop Dean." The "Stop Dean" title could be conferred on February 3. Of the five primaries and two caucuses to be held then, several are in conservative states (South Carolina, Oklahoma, and Arizona), where Dean is expected to be weaker.

The biggest prize on February 3 is South Carolina, whose Democratic primary is the first with a significant black vote. So Al Sharpton and Carol Moseley Braun are expected to do well there, but so is Edwards, who is originally from South Carolina and represents North Carolina in the Senate. Lieberman, a relatively conservative Democrat, is also expected to be competitive there, and so is Clark, a Southerner and a retired general. And if Gephardt survives Iowa, his criticism of free trade might resonate in South Carolina because of its battered textile industry. Gephardt also has the support of Rep. James Clyburn, the state's only black member of Congress.

In other words, anything could happen in South Carolina. Gephardt, Sharpton, Braun, Edwards, Clark, and Lieberman could split most of the South Carolina vote, in which case Dean might come in first. Such a situation would be so unexpected, it could effectively end the contest for the nomination by showing that front-runner Dean does well even in the South.

If someone other than Dean wins South Carolina—or Arizona or Oklahoma—on February 3, that candidate might clinch the "Stop Dean" title. And the stage would be set for a brutal monthlong slugfest between Dean and Stop Dean. Other candidates would be expected to drop out, and in each case, the question would be: Does he or she endorse Dean, try to stop Dean, or stay neutral?

For the rest of February, the battle will rage across seven statewide caucuses and four primaries. First up are the Michigan caucuses, on February 7. Will a major Midwestern battleground state go for Dean? His chances may be good, thanks to Michigan rules that allow Internet voting. Dean's supporters are well wired.

Then come the Virginia and Tennessee primaries, on February 10. Will Al Gore deliver his home state of Tennessee for Dean? Or will Democrats in those two Southern states bolster the Stop Dean cause? If Clark wins South Carolina, then Virginia and Tennessee could generate real momentum for him.

The Democratic race comes to a head on Super Tuesday, March 2, when nine primaries take place, including in California, New York, Ohio, and Georgia. That day is likely to be the final showdown between Dean and Stop Dean. On March 2, only one of their campaigns is expected to come out alive.

Of course, candidates and their spin doctors are always trying to lower expectations so they can boast that they've done better than "expected." And new polls keep coming out, so expectations can change in a flash. That's the problem with running against "expected." He's a candidate whose positions keep changing.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.