(Other People's) Tea Leaves

There are two basic numerical questions in the Indiana and North Carolina primaries: who wins the most votes, and how many more delegates does the winner get?

Realistically, Obama wins a draw, because the largest remaining bunch of delegates will be gone by tomorrow's results. Clinton will likely go on if she gets a split in popular vote wins. But her chances of getting the nomination diminish.

So on to the tea-leavers. This post will be updated throughout Monday and early Tuesday with whatever new predictions there are.

Final SUSA polls show Obama up by 5 in NC. Key comment: There is no foreseeable outcome in North Carolina, regardless of which candidate wins the popular vote, where one candidate collects significantly more convention delegates than the other. CQ Politics concurs--on the district level, they see Obama gaining only 3 more delegates. However, the blogger Poblano, whose PA predictions based on math models I don't pretend to understand were just about dead-on, predicts a much bigger victory for Obama, of 17 points and a 66-49 spread in delegates. He emphasizes that polls significantly underestimated Obama's totals in southern states with big African American populations.

Another wonky online analyst with a good record, PsiFighter37, touts Obama winning 63 delegates in North Carolina to Clinton's 52. He also sees a double digit popular vote margin.

Poblano's analysis of Indiana is a Clinton win by 51-49%, with an even split in delegates. PsiFighter37 says Clinton wins 4 more delegates, with a a Clinton advantage of 6.4% in the popular vote. Most late polls show Clinton ahead, although the margins vary from 6 points to more than 10. Zogby is holding firm with a slight margin for Obama.

And here's a summary of where things stand on Wednesday. The major point: there will be more undeclared super-delegates than delegates at stake in the final contests. Chuck Todd on Keith underlined this point by saying that tomorrow is the last really significant election day of the primaries, and unless Clinton wins North Carolina and Indiana both, she has almost no chance of getting the nomination, either by delegates won or by convincing super-delegates that the momentum has shifted to her.

Update1: Latest PPP poll in Indiana shows Hillary ahead but it's Obama who has the last minute momentum.

Update2: Al G. at The Field has posted his predictions. He gives Clinton a big 10.8 win in Indiana gaining 8 more delegates, and Obama a 6.8% victory in North Carolina, gaining 7 more delegates. So Clinton ends up with one delegate more--far, far fewer than she needs to change the dynamics of the race. Al obviously feels that Clinton's pandering, the TV obsession with Rev. Wright and the rabid right interest in derailing Obama are going to take their toll. It's also worth mentioning that he turned out to overestimate Obama's vote in PA, so there may be a little compensation in this prediction. He also points out, as I did at the top, that a draw is a win for Obama, at least technically.

Update3: Booman at Booman Tribune has posted his Indiana district by district predictions. The outcome he sees is Clinton winning the popular by 51-49 and gaining 4 delegates. Meanwhile, Election Projection has its North Carolina projections: Obama 58.5% to Clinton 39.7%, giving Obama 15 more delegates.

So these are numbers analyses. I've also obsessively read a number of campaign reports. I'm always looking for signs favorable to Obama, and two particular ones stick in my mind: for the recent Jackson-Jefferson Day dinner in North Carolina, equal numbers of tickets were sold to Obama and Clinton supporters. However, fewer--apparently far fewer--Clinton people showed up. The immediate consequence was that Clinton's speech was swamped by Obama chants, but the greater indication is--what? Bill Clinton has been hitting the rural areas, and this could pay off in votes. But will that do much to dampen enthusiasm in the bigger places? Also, North Carolina saw massive early voting, with over 40% by blacks, and--one can surmise--a lot of the white vote by students, who may have voted where they go to school even if they won't be there on election day. Big schools in Indiana won't be in session, so students in Bloomington, etc., had to early vote. I haven't seen numbers on that.

In Indiana, the momentum seemed to be going towards Hillary until this weekend, and if it continues moving toward Obama, that 3% a lot of these folks are giving as Hillary's margin could swing. Here's what tantalized me: a reporter's observation that Gary is solid Obama, not leaning that way. Demographically, that doesn't compute. But could a reporter's instincts suggest something out of the ordinary--white working class voters going for Obama?

I was glad to see a story saying that the Obama people were energized, motivated and optimistic in Indiana. That's his secret weapon there: the Obama faithful from Illinois who can work with mayors backing Obama and counteract the Evan Bayh regulars.

Okay, I'll say it: I'm looking for a double digit win in North Carolina, with double digit delegate gains, and a single digit loss or a victory in Indiana. I have this feeling he's turned the corner. (Okay, I've had it before.)

Update5: the final update in this thread is the Kos predictions: he has Clinton in Indiana 51.1 to 48.9 (which would mean it will be the late late show before the networks can name the winner) and Obama by 12 pts. in North Carolina (which means the cable gab will be about Obama's resurgence.) But I thought it was most appropriate to end this thread with Kos explaining his "system": "If my system sounds stupid, it's because it is. They all are. Ultimately, no one predicts every race accurately, not even the pollsters who actually ask people who they're going to vote for." Amen.