I figured Zogby would "adjust" his final numbers in line with the other pollsters and "discover" a late "move" toward Clinton in the last night of polling in Indiana. But the opposite seems to have happened.

So once again, it's Zogby v. SUSA. In California, SUSA cleaned Zogby's clock. Is this Zogby's revenge? Or is it California redux?

I think he has NC about right, and yes, I think Obama's strength among early voters (according to every pollster) makes it tough for Clinton to make very close.

The opposite seems to be true in Indiana. SUSA shows Hillary doing better among early voters there, however early voters only account for 3% of the their sample (so the margin of error for that subsample would be so high as to make the numbers worthless). I think Obama may be able to keep it within 5.

Zogby appears to show Obama doing better among white voters than some other pollsters. On the other hand, I think Obama will get more of the AA vote than Zogby's numbers indicate (79% is what I believe he shows). Who really knows. It's all about turnout models when you have states like these where candidates each have distinctive demographic and geographic strongholds.

It's Zogby.  I could place index cards showing various margins in a circle and have one of my pet turtles crawl over one of the cards, and that would be more scientific than his polls.

Wow, that is a not expected.  Zogby's final night polls have been extremely accurate as of late.  Did Gas Tax blow up on her?

You predict 10% of voters in NC will vote for someone else and even higher in IN???

That's a fair prediction except that your numbers only add up to 90% and 87% respectively!  ;-)

I'm not buying it, considering the other polls, especally the SUSA one, showing a formidable lead for Hillary.

I'll stick with +8 in Indiana and -8 in NC.

Edwards must have relaunched his campaign

Let's hope Zogby has indeed learned from Super Tuesday and keeps his great Post-Super Tuesday record up.

All the Indiana polling appears to underestimate the AA turnout. SUSA and IA have it at 10%, while PPP is at 12% and Zogby at 11%. The overall AA population in IN is roughly 10%, which would suggest something more like 15% for a dem primary in a reddish state.

This works both ways, however. Most of these polls also have the female number too low.

Moreover, age and geographic distribution is important, given the distinctive bases for each candidate.

The accuracy of these polls depends ultimately on the turnout model, and frankly, I don't like the turnout models for any of them. I imagine that they are based on past IN primaries, but none of those really matter (and neither involved a female or AA candidate).

Although I'd love it is Zogby was right, I think he's off his rocker.

NC should go to Obama by 8
IN will go to Clinton by 6

Zogby may be overestimating AA turnout in Indiana.  The supreme court vote on the voter ID law will certainly depress turnout there as well as among new voters.  Add that to the fact that Indiana is an open primary and Rush Limbaugh has asked all Republicans to vote for Hillary as part of Operation Chaos.

I think though this is the first time in a while (WI really) Obama has had a better closing weekend than she has.  The regression of his "surge" has not taken place over the weekend as it did in OH, TX and PA.