"If she is unable to win by a solid double-digit margin come April 22, I do believe that she is really going to have to reconsider whether she is actually able to secure the Democratic nomination."

We can only hope she has that much sense and grace.

You forgot to mention the new Insider Advantage poll that has her up by 10

The interesting aspect about these polls is that Obama never even comes close to breaking out of the low 40's.

All the support Clinton had seems to have been transferred to undecided.

Perhaps it is unprecedented because this is an unprecedented campaign in Pennsylvania?  A potentially decisive election in a very expensive media state?

There is a new Insider advantage poll out that shows her gaining 8 points from their last poll , she now leads by 10 points .

Note they didn't push leaners , you still have 13% undecided , so its not out of the realm of possibility that survey usa is right.

Here is the link :

http://www.southernpoliticalreport.com/

You can update your information based on that , if you want.

yeah, that new poll with +10 lead for Clinton kinda goes against what you're stating in the diary (serious tightening).  I think we're seeing a difference in sampling that's causing a lot of the variance.  I do believe there's a tightening- but i think that has happened in most of the recent primaries that Clinton has won.  In the end this could be bad for Obama, as I commented before.  If he loses in the double digits, it's going to look as though no amount of money will win over the "Reagan Dems" to his campaign.  And by that I mean white people in the battleground states (let's just be real here).  He is spending money like Bush in Iraq- and this could backfire on him.  Just my opinion.

PPP shows her trending up.

look at the movement in the ppp poll!

5 point swing!

same as survey usa swing from their last to latest poll.

Yet another BS attempt to set the goalposts at an impossible distance.  If Clinton wins Pennsylvania, Indiana, West Virginia, and Kentucky, the momentum will shift and the real question will be, not how much did she win by in Pennsylvania, but 'how did Obama do in North Carolina and Oregon?'  If his numbers are collapsing, the "who is more electable in November" question will be back on the front burner.

It would be a big stretch at this point to call it a tie. Only ARG currently shows a tie, most polls show +3 to +8 or so for Clinton. That may be technically "within the margin of error" on the basis of each individual poll, but taken together represent a much larger sample that shows a fairly consistent Clinton lead (though obviously greatly reduced from a few weeks ago.)

Remember too that PA is a machine state, and the entire Democratic party machine is behind Clinton in this race, aside from the Casey family. That's a built in advantage of a few points that might not be reflected in polling.

". . .so much so that Obama might actually have a valid shot at winning."

Just talked to friends who were canvassing in the Philly area for Obama and the stuff they heard. . .wow.  There are plenty of legitimate reasons not to vote for Obama - but the overt racial slurs/comments that a surprisingly large number of voters made while talking to canvassers was shocking.  

Just got to have thick skin and move on right?  There are many places that Obama will simply not be able to win in without that backing of the Democratic establishment, and PA is one of them.

This is my first campaign canvassing, and you definitely experience a lot.  From DC, where voters didn't want to vote for Obama because they thought he would be assassinated to PA where voters refer to him as a *igger.  Sobering to say the least.

I think the margin of error is nonsense as it does not include adequate turnout models.

No, more and more it's looking like the contest in Pennsylvania is tightening up quite a bit -- so much so that Obama might actually have a valid shot at winning.

"Just talked to friends who were canvassing in the Philly area for Obama and the stuff they heard. . .wow.  There are plenty of legitimate reasons not to vote for Obama - but the overt racial slurs/comments that a surprisingly large number of voters made while talking to canvassers was shocking.  "

Sorry you had a negative experience in PA. I've been canvassing hundreds of families in NE Philadelphia over the past weekends and have not heard one single racial slur. Not even a hint of one, and I am very attuned to such things.

I have heard a great deal of positive and enthusiastic support for Hillary Clinton, much of it around the health care issue. People have been very hopeful that she will be able to improve the situation on health care, and seem eager to trust that she cares about them.

Many women support her and are thrilled at the idea of her as our first women president. Many want to do more to help her, and regret that they have child rearing and family responsibilities that prevent them from volunteering. It is very clear that Clinton has very sincere support in PA for a variety of reasons.

I wouldn't attribute PA voters' support of Clinton to racism, there are too many positive reasons to vote for her.

Not that there aren't racist voters in every state, just as there are misogynist voters in every state who will not ever vote for a woman for president, no matter how qualified. But tarring a "large number of voters" supporting your opponent as bigots of one sort or another is ambiguous at best. It's an insult to PA voters.

Barack Obama is a political descendent of Karl Marx, who was Christian in his youth and must have felt the spirit of the earliest Christian community in Jerusalem as ideal

Today's Marxist-oriented news media and today's Marxist public schools will continue propagandizing, but so-called "progressives" like Barack Obama have been successful in implementing the majority of this manifesto in most countries of the world. At what price? In communist countries, tens of millions of non-believers in Marxism have been slaughtered in cold blood or left to rot in political prisons.

Clinton is gonna win Penn. by 10+ IMHO. It seems to me that Obama can't beat Hillary there and I'm an Obamabot.

There's an inordinate amount of attention being paid to the margin and not enough focus placed on the percentages themselves.  The fact that Obama's only three points behind Hillary doesn't really matter considering he's still a full 7 points from 50% and thus from ultimate victory.  Clinton is losing hard support, but as SUSA shows the undecideds, when pushed, still lean her way.  I do think SUSA was off on Obama's number... I think it's safe to assume that he's within the 43-45% range right now, but not a single poll has shown him above that level.  He may be bouncing against his ceiling, somewhere right around where a double digit loss becomes a single digit loss.  That's good news for him so long as expectations don't get raised to ridiculous levels, and suggesting that Obama has a real shot of winning based on the polling information we have right now is ridiculous.  

You know, it would be nice if Beeton or Armstrong would write posts like this.

I think the weightings are off in other polls for PA.

You don't even need to follow polls this far out. Just follow the money. Obama spent the last two weeks bombarding PA with something like three times the ad volume as Clinton. So, naturally, he moved up. Clinton has now begun to even it out a bit, so she's moving. This is all crap movement. Wait until the last weekend and then look where the daily numbers are going. That will tell you whether Clinton wins by double digits or it's tight (or even an Obama win).

Ohio was double digits..the result

                              Obama   Clinton
Ohio    03/04    141         66    75

There's a new SUSA poll showing Clinton up by 20 in Puerto Rico.

Here's a video  of Bill's visit to the island.

Yeah but could she carry PR in the GE?

I'm not saying discussion is bad but why is so much energy expended debating poll results? Polls are notoriously bad predictors and change so much from day to day that it's really a poor investment of time to give them too much thought.

Even if they were good predictors, what good does it do to wonder what will happen on the day of the vote? You can't change it unless you are a PA voter and then it's just your one vote. It's going to be what it's going to be. The success of the Democratic nominee will be much better served thinking about how to beat McCain and what we can do to serve that purpose.

I know everybody loves a good prediction and I'm no different. But I tend to only get in the prediction game when I can gain something from it. I don't see a gain in this. We will all know on April 22 exactly what the split is and this a real discussion can begin over what the results mean for the nomination process since that has real meaning in terms of winning in November.

it's got to be close. why else would lanny davis write that wretched op-ed in the wall street journal?

Like Ohio, I can see Hill by 10!

...who wins the primary, I have to ask.

Why is this good news?  She was beating Obama by 20 points a month ago.  And now there's cheering that she's "up within margin of error"?

I can't understand how this could be good.  She can't afford to win PA by "margin of error".  she needs to win it and every other state big.

It is wishful thinking by Obama suporters that it is  very close or tie.

I will bet that Obama is going to loose by more than  25 points against Hillary in PA.

In spite of spending enormous amount of money and even with Obamamaniac press, Obama is still NOT leading in a single poll.

Obama is trailing even in lousy polls like Zogby, PPP and ARG.

McCain is going to pick Tom Ridge as VP candidate and Obama is finished for the General Election.