In the looming battle for superdelegates, all kinds of arguments have and will be put forward for how they should decide who to support.

One that has been embraced by a lot of people is that superdelegates, for the good of the party, must subordinate themselves to the will of the people, as expressed in pledged delegate counts. If the party bosses foist on us as nominee the person with fewer pledged delegates, Democrats will feel bitter and disenfranchised. They may not go so far as to vote for McCain, but they may go so far as to not vote. Or not volunteer to work on the campaign and get-out-the-vote efforts. Low turnout and lack of election enthusiasm might just put McCain in the White House. For some reason, this is the argument embraced by the Obama campaign:

“It is very difficult to see any scenario that Hillary Clinton would get the nomination in a way that doesn’t rip the party apart,” said Wisconsin Gov. Jim Doyle, an Obama supporter. “I think that it would be a terrible mistake for the Democrats to not accept the will of the people who have turned out in primaries and caucuses.”

One that has hardly been embraced by anyone at all yet is that the will of the people may also be judged by popular vote totals, not just by the pledged delegate count. But look for this to loom larger if Clinton takes the popular vote lead (which she whittled down from 4.6% to 2.3% in one voting day last Tuesday). Or even if she comes close to taking the lead.

Then, there is the whole electability argument. Superdelegates have a duty to vote for the candidate who they think will do better in November. Not just a candidate who will beat McCain, but the candidate who will produce a stronger Democratic mandate, the candidate whose coat-tails others can ride to victory in close House and Senate races.

Just yesterday, this view was advanced feelingly on TV by Bob Casey, the junior senator for Pennsylvania, on MSNBC and ex-Senate Majority Leader Tom Daschle, on The Daily Show. It was also echoed (but without the “mandate” angle) by the Clinton campaign, which stresses that Clinton has beaten Obama in key November battlegrounds:

Clinton spokesman Doug Hattaway said Obama’s lead in pledged delegates is “hardly a mandate.”

“Some superdelegates will go with (the) pledged delegate count, but many will go with the candidate they think can win,” Hattaway said. “We have a very compelling case to make on that front, given that we’re winning general election swing states, must-win states and must-win constituencies.”

In this post, I don’t really want to get into the respective merits of these three arguments (support the pledged delegate winner, support the popular vote winner, support the one who’ll do better in November). I want to focus on the issue of how superdelegates can and should judge electability.

One simple, easy way is to go by the head-to-head matchup polls that tell us how Obama and Clinton would do against McCain. The latest national poll (by The Washington Post and ABC News) shows Obama beating McCain by 12 percentage points, and Clinton beating him by 6 points. Obama doesn’t just do better today, he has consistently done better against McCain in these national poll match-ups than Hillary has. Here’s Chris Cillizza of The Washington Post on February 11:

As evidence, the memo cites recent polls from Time, CNN, Cook Political Report, Post/ABC, Fox News and Rasmussen — all of which show Obama running slightly to considerably stronger than Clinton in hypothetical matchups against McCain. A new survey out today — conducted by the Associated Press — affirms that idea with Obama leading McCain, 48 percent to 42 percent, while Clinton leads McCain, 46 percent to 45 percent.

The Clinton campaign is clearly encouraging superdelegates to think in “more sophisticated” ways. They would like them to engage in a more intangible calculus, look at who won key contests, and extrapolate from the quality of those victories. Doing this yourself is impossibly hard, but this is pretty much what is captured by state-by-state polls (which, of course, are much more relevant to a presidential election than a national poll is). SurveyUSA has just come out with a poll that shows Hillary beating McCain, 276 electoral votes to 262. Obama does better: 280 to 258. (There’s quite a difference in which states Obama and Clinton win, by the way. Interestingly, Clinton loses Michigan but Obama wins it. Florida is the reverse: Clintons wins, Obama loses.)

So the state-by-state polls show a less dramatic edge for Obama, but he still comes out doing better than Hillary.

In my mind, though, the real key is not who would do better against McCain if the general election were held today (which is what the polls measure, ultimately), but who is more likely to do better in November.

That takes us right back into the realm of touchy-feely, intangible calculus. You don’t go just by what the relevant head-to-head polls are saying today; you make your own personal assessment of what they are likely to say tomorrow. Under the electability argument, that’s the judgment that a superdelegate is duty-bound to make.

And that judgment need not be made mechanically. Not in mechanical response to knee jerk talking points from the two campaigns. Not in mechanical response to what the polls show today.

For what it’s worth, if I were a superdelegate, and if I bought into the electability argument, then here’s how I would be thinking today.

The conventional explanation for why Obama does better than Clinton against McCain is that independents prefer Obama to McCain, but show them McCain versus Clinton, and too many of them shift to McCain.

It is my considered opinion that unless these Obama-to-McCain independents are from the fabled Hillary-hater brigade, they can only prefer McCain to Clinton today because they actually know very little about McCain’s positions on various things. Just as Rudee‘s so-called candidacy progressively fell apart the more voters learned about him, so will these Obama-to-McCain independents push themselves further and further away from McCain the more they learn about him.

Not only has he been spectacularly wrong about the Iraq war from the beginning (and spectacularly dishonest in retrospect about his position on the Iraq war), not only does he have no clue at all about the economy, not only does he mindlessly embrace Bush’s discredited, bankrupt policies on both Iraq and the economy (right down to the voodoo tax-cut philosophy), but this man has also shamelessly prostituted himself for the presidency by abandoning almost every principled position he once held (see here for an excellent summary). There is no way any independent who would pick Obama over McCain and who actually looks at McCain’s record of being wrong and of shameless pandering — or has it brandished in their face by a Democratic candidate — is going to end up picking McCain over Hillary.

So I don’t really buy the proposition that what the polls (whether national or state) show today as Obama’s electability advantage is going to last into November. I believe that Clinton’s numbers against McCain will keep moving closer and closer to Obama’s numbers against McCain, as more and more of today’s Obama-to-McCain independents decide they are really Obama-to-Clinton independents.

In short, I really don’t think there is an electability differential. Meaning that there wouldn’t be by the time November rolls around.