Put yourself into the mind and mood of an uncommitted Democratic superdelegate. You are getting inducements and pressure from both the Clinton and Obama campaigns — including direct appeals from the candidates themselves that range from the intellectual ("I will be the better candidate in the fall"), to the seductive ("You will be a player in my administration"), to the vindictive ("Get on the train now or there won't be a seat for you later"). But you are also juggling a set of other intense pressures: Who did your state or district support, and would you erode your own standing if you went the other way? Who really could win the White House, based on the nomination battle so far? Who would be best for "down-ticket" candidates — perhaps including yourself? And what happens if you back the wrong horse?

Theoretically, your responsibility as a superdelegate — one of the 20 percent of Democratic convention delegates who receive a nominating vote because of a seat on the Democratic National Committee, or as a Democratic elected official or a party dignitary — is to do what is best for the party. That is, pick a winner and heal all the wounds before November.

But how to square that circle?

If you cast aside Barack Obama, who's leading the national popular vote and in elected delegates, in favor of Hillary Rodham Clinton, who you may believe could lure the most swing voters in the general election, it would almost certainly create a firestorm so great that it would create a new Wikipedia entry under "Pyrrhic victory."

On top of that, you know the Florida-Michigan problem is festering. Both states are barred from sending delegates to the Denver convention because they held their primaries too early. If some way isn't found to represent these states in the process, the Democratic voters there could sour on the whole party come November. Probably the best way to take care of it is for you and your fellow supers to settle on a candidate before August. If the essential decision isn't in the balance in Denver, then no one will fight the inclusion of Florida and Michigan. Meanwhile, each day that goes by deepens the divisions along age, class, race and gender lines, with animosities whipped to a frenzy by blogs and cable TV talkfests.

Oy.

Of a total of 795 Democratic superdelegates, there are just fewer than 300 who still haven't made up their minds, according to a running tally at www.realclearpolitics.com. Al Gore is one of them, and he points out that few covet the role of deal maker-kingmaker. "No one wants to be an umpire," he's reported to have said privately.

Party insiders claim that a majority of the undecided were poised to end the contest and their own pivotal role on March 5, the day after the Texas primary, if Obama had given them the excuse and won by a single vote (instead, he lost by just fewer than 100,000 — ironically, with well more than 100,000 Republicans crossing over to vote for Clinton). Had Obama won in Pennsylvania, or come within a couple of percentage points, that too could have sealed the deal for most of the remaining superdelegates.

Why the lean toward Obama? Pragmatism. Obama has an insurmountable edge in terms of elected delegates (156) and a robust lead in the popular vote (roughly 600,000, excluding pesky Florida and Michigan). How do you deny the nomination to an African-American candidate who wins the most delegates and the most votes?

So why then haven't the uncommitted superdelegates ended their agony and opted for publicly declared support? First, there has been no knockout blow to Clinton. Without it, declaring for Obama could infuriate a solid core of the Clinton base, creating the prospect of another kind of Pyrrhic victory. Second, Obama's failure to deliver that knockout has given many of the supers pause — a desire to wait and see if another shoe drops. Is there a scandal or revelation strong enough to knock the front-runner permanently off course? Will there be serious new fallout from the suddenly ubiquitous Rev. Jeremiah A. Wright Jr.? Will Obama's problems with downscale white voters worsen?

Of course, the tough and experienced Clinton team is doing everything it can to fan the flames of those doubts. For Clinton, though, the stark reality is that her gritty performance over the past two months is not enough. In the absence of a knockout blow to Obama, or of overwhelming polling evidence that she is a stronger general election candidate than he (which so far is not there), she has limited sway over the remaining superdelegates.

At a bare minimum, Clinton needs to net another 300,000 votes over her rival in the remaining contests to make a reasonable case that she has equaled or exceeded him in popular votes by including her edge in the noncontest in Florida. (Given the dwindling number of remaining contests, she cannot afford to "lose" many of the 200,000 votes she picked up over Obama in Pennsylvania.) That adds up to Clinton achieving at least a low-single-digit-margin loss in North Carolina and a significant win in Indiana.

As the front-runner, Obama is playing it cool, although he is still using his spare moments to call superdelegates and chat. If Obama wins Indiana, odds are that a pent-up flood of supers finally will endorse him on May 7, signaling an early end to the process. If he holds Clinton's edge in Indiana to somewhere close to 2 percent and wins North Carolina by anywhere close to the 15 percent margin he holds in the polls, odds are that a trickle of superdelegate Obama endorsements will become a steady stream, resulting in a slower May decision. If neither of these scenarios pan out, the supers will try mightily to resolve the issue on or after the last active voting on June 3, to keep their convention from careening out of control. But they need a good reason to end their torture. Ambiguous results, damaging enough to Obama to keep him from closure but not definitive enough to move these reluctant power brokers to the Clinton camp, are their recurring nightmare.

Norman Ornstein is a political scientist and a resident scholar at the American Enterprise Institute. He wrote this commentary for the Los Angeles Times.