Superdelegate redux

Blogger Dylan Loewe, in a post titled "Why Obama is Closer to the Nomination Than You Think," argues convincingly that, in the end, neither rogue superdelegates nor the Florida/Michigan controversy will decide who wins the Democratic nomination; the winner will be whoever gets the most pledged delegates, even though the pledged delegates alone won't get the winner anywhere close to a majority. The superdelegates, Loewe says, are too wimpy to buck the will of the pledged delegates -- and the will of the pledged delegates will similarly control who gets on the Credentials Committee, and thus whether the Florida and Michigan delegations get seated. (Hat tip: Noam Scheiber.)

I think Loewe is probably right, provided the leading candidate's pledged delegate advantage is meaningful -- say, over 100 delegates or so. If that's the case, the superdelegates will rally around the leader for the sake of party self-preservation. But if the leader's delegate edge is in double digits only, I think all bets are off, in part because a lot of these superdelegates will be making their decisions while the national delegate estimates are still maddeningly in flux (so the identity of the "leader" may be less than clear) and in part because Hillary will then be able to wield the "popular vote" argument as a potentially effective counter to Obama's "I got more delegates" argument. If the final pledged delegate count is, say, 1,640 to 1,587, then it really will be chaos. But if it's more like 1,700 to 1,527, then, yes, the nominee will be the pledged delegate winner, notwithstanding that 1,700 is well short of the 2,025 needed to nominate.

P.S. With regard to those "maddeningly in flux" delegate counts: the media seriously, seriously needs to start consistently and clearly separating superdelegates from pledged delegates in its counts. To view the numbers primarily as an undifferentiated mish-mash is so confusing as to be almost worse than worthless. A "committed" superdelegate is not the same thing as a pledged delegate, for the simple reason that the superdelegate can change his or her mind, as some already have and many ultimately will. I'm not saying the superdelegate tally isn't important, but it is separate from the pledged-delegate tally, and it needs to be regarded as such.

If I were a major media organization with a "decision desk" and a staff of analysts figuring out delegate counts, I would publish a breakdown of delegates that would explicitly include, in separate columns, the following clearly delineated categories:

* Pledged delegates that have been allocated

* Pledged delegates that are at an intermediate stage in the allocation process, but whose ultimate loyalties can be predicted with reasonable certainty (e.g., the "delegates" in many of the caucus states, where what have actually been elected thus far are delegates to county conventions, who will in turn elect state delegates, who will elect national delegates)

* Pledged delegates who are tied to a primary or caucus that has already occurred, but whose loyalty cannot yet be predicted with any reasonably degree of certainty (due to an unpredictable multi-stage process, a too-close-to-call vote, etc.) -- and I would break these down, if possible, into "Lean Hillary," "Lean Obama," "Total Tossup," etc.

* Pledged delegates who are tied to a primary or caucus that hasn't occurred yet

* Known superdelegates who have made an endorsement

* Known superdelegates who have not made an endorsement

* Superdelegates whose identities are not yet known, but whose loyalties can be predicted with reasonable certainty (the UADs)

* Superdelegates whose identities and loyalties are unknown

I will give $20 to the first person who creates a webpage that breaks the delegates down into these categories and then tracks them continuously. Heck, $30. :)

Comments

[Condor]

Loewe makes a good point. But it seems to me that it's a little too simple to say that the superdelegates won't buck the will of the pledged delegates. Suppose Obama wins by 100 pledged delegates. Meanwhile Hillary has 150 more "committed" superdelegates. Do the uncommitted superdelegates break for Obama or Hillary? Now it could be that the uncommitteds won't buck the will of the people. But let's suppose the uncommitteds break evenly. This would mean that in order for pledged delegates not to buck the will of the people, they would have to change their votes. But rather than following the principle that "the will of the people is best" it seems more likely that they would follow the principle "make the most stable, least chaotic, decision possible." This would mean sticking with their original pledge.

[Joe Loy]

" If the final pledged delegate count is, say, 1,640 to 1,587, then it really will be chaos. But if it's more like 1,700 to 1,527, then, yes, the nominee will be the pledged delegate winner, notwithstanding that 1,700 is well short of the 2,025 needed to nominate."

A bit of Dubiously-relevant precedental Perspective. 1976 Republican National Convention rollcall results:

U.S. President Gerald R. Ford...............1,187
Former CA Governor Ronald Reagan....1,070

Granted, there were few-to-No official "superdelegates" as Such. But there Were informally-"uncommitted" delegates, up until even Past the 11th Hour. / Desperately, RR tried to Woo them :} by absurdly naming a prominent out-&-out liberal, Sen. Richard D. Schweicker R-PA, as his Runningmate slightly in Advance of the convention. / This did not Work :}. In fact, it Backfired. The Fordies (including the massive Pennsylvania contingent thereof) were Unmoved, apart from the Part where they went ROTFLTheirA'sO :>. The Reaganisitas, by contrast, were Appalled. Yer Man had to spend hours on the Horn pleading with them to Stay the Course.

But the point (if Any :) is: Conventions CAN be Close.

[Derek]

I think the article's a bit too simplistic. I think that you're far overestimating the "collective action" of the superdelegates. There's no question that there are, and will be, defections here and there. But I think it's unlikely that there will be a massive transfer of "committed" superdelegates near the end. A lot of these are rabid Clinton supporters (some just picking the winner). Some don't want to spurn Clinton, then find out that FL & MI are counted and she's back in.

For now, most (not all) "uncommited" superdelegates have indicated that they're (a) waiting until the end, or (b) waiting until their state comes in. They're people who don't expect a "favor" in the spoils system.

But why are people still endorsing Clinton and Obama as of yesterday? I thought it was "too close"! It's because, I think, they expect that endorsing before it's inevitable means that they'll be more likely candidates for a "plum" administration job. If you wait until it's inevitable, there's no spoils for that; the administration will snub you. If you pick early and win, then you're a "loyal" supporter.

About 50% of supers have committed at this point. About 50% have not. They're still cutting slightly in Clinton's favor, but more recent commits have broken for Obama. Nevertheless, the temptation for supers to pick the winner now rather than get stuck just "doing the duty" of "voting for the popular vote winner" continues to shrink, overwhelmingly for, I would argue, sensible self-interested reasons.

[dcl]

I don't have the access to data that I would need to do that, sorry.

[dcl]

hmm, brendan, does this help:
http://demconwatch.blogspot.com/2008/02/ultimate-delegate-tracker.html