Updated: And now for the Clinton comeback?

The first Super Tuesday checked Obama's momentum--but then he recovered with 11 straight wins in the following contests. That second remarkable surge had his campaign hoping for a knockout blow today--and even had many commentators calling for Hillary to withdraw. The Clinton campaign, playing the expectations game with some success, is now getting ready to deem anything but a clean sweep by Obama in the second Super Tuesday a setback for him. On the face of it, that seems absurd, but if Hillary wins comfortably in Ohio, and holds Obama to a draw in Texas--which is what the late polls are pointing to--the momentum will be back with her going into Pennsylvania, and this thing is by no means over.

Note that what follows is laid out in much more detail by the excellent Jay Cost at Real Clear Politics. This is just an abridged version of his analysis here and here. His line seems dead right to me.

The ceaseless focus on the elected delegates exaggerates the difficulty Hillary faces in turning this around. Unless Obama buries Hillary today--and the polls say that won't happen--this race will go on, and it will turn in the end not on the ordinary delegates, but on the superdelegates. The question is, how will they decide which candidate to back? There are several scenarios, but the most likely is that they will be swayed by the popular vote. If Hillary can get ahead of Obama in the popular vote, she will have a strong moral claim to the support of the superdelegates. With the race for ordinary delegates inconclusive whatever happens in the remaining primaries, competition for the popular vote is the race that matters.

As of this morning, Hillary was behind roughly 52-48 in the popular vote. With the states remaining, she can still turn that around. Essentially she needs big winning margins--but not huge winning margins--in Texas, Ohio and Pennsylvania combined. That might not put her ahead in the elected-delegate count, but if she ends up ahead in the popular vote she can hope to get the superdelegates on side. The necessary winning thresholds for a lead in the popular vote depend on whether you take Michigan and Florida into account. Without those states (disqualified, for now, from the elected-delegate count) Hillary needs to win more than 55 percent of the popular vote, starting today, in all the remaining states taken together. If you include Florida (where Obama was at least on the ballot, unlike in Michigan) she needs to win more than 53 percent. If you include both Florida and Michigan (which would seem indefensible, though that will not trouble Hillary) the requirement shrinks to 52 percent. Reaching even the most demanding of those thresholds is hardly impossible.

Obama is still the favorite--and will remain the favorite unless Hillary crushes him today (in absolute terms, not relative to expectations). But if she does enough to stop the Obama momentum again, she will have a fighting chance of emerging the eventual winner. And, as Hillary never tires of saying, she is nothing if not a fighter.


This piece by Jonathan Alter goes through the delegate arithmetic. On the assumptions he describes--which are indeed quite favorable to Hillary--she can't catch Obama on the elected-delegate count. But consider this paragraph:

So no matter how you cut it, Obama will almost certainly end the primaries with a pledged-delegate lead, courtesy of all those landslides in February. Hillary would then have to convince the uncommitted superdelegates to reverse the will of the people. Even coming off a big Hillary winning streak, few if any superdelegates will be inclined to do so. For politicians to upend what the voters have decided might be a tad, well, suicidal.

"Reverse the will of the people." But what if these assumptions put Hillary ahead in the popular vote? I can't run those numbers, and they depend in any case on further assumptions about turn-out. (Jay, help me.) But I'm guessing that the scenario Alter describes could be enough to put Hillary in front on the popular vote (especially if you include Florida). In which case, it would be Obama, not her, who could be accused of using the superdelegates to reverse the will of the people.

This is still unlikely, because Alter's assumptions do, as he says, lean heavily in Hillary's favor. But the point is, she can lose the elected-delegate count and win the popular vote--and that, in the end, may be the vote that matters most.