I'm amused by the use of the italics in your headline, Eric.

Don't believe it Obama supporters. Yes, he's got the momentum on his side, but he's definitely still losing. Keep working like he was down 10-15 points.

It is an outlier, but it isn't as though their past polls have favored Obama, and all the polls show movement, so there really might be something here. If these numbers are even close to accurate Hillary is in big trouble.

And really this isn't that far from Rasmussen's which put Obama 5 down...so I'd say don't dismiss it.

Even as an Obama supporter I cant really belive this is true. PPP has had a solid record this cycle but a 26 point swing??

On the other hand, it probably does mean hes closing the gap, meaning a Clinton blowout is becoming less likely

WOW!!! Even if this poll is an outlier, this will sure get some heads turned. He has three weeks to catch up. he almost did it in Ohio and TX and he has a good shot to do it in PA.

After the Jerimiah Wright flap i was so deflated cause i thought he wouldnt recover, but that flap made BO stronger. I am trying so hard to temper my excitement, but it seems that HRC wont get that blow out that she so desperately needs in PA

"this is a huge outlier"
BS... The average of the last 3 PA polls is near +5 for Hillary. Finding a poll showing them nearly tied is not an "outlier" - it just confirms the Rasmussen poll.

The news barrage of Wright, Obama's race speech, and the Bosnia snafu has sent polls into an amusing spasm over the past couple weeks.

I shudder to think what Zogby's reading on PA would be.

This is absurd, it honestly looks like they just polled Philadelphia, then decided those results were good enough. Looking forward to pollster's take on these numbers.

It raises a red flag that PPP was also the firm that had a huge swing in the NC poll just last week as well. What came out was that PPP had changed the pool of voters it was sampling (expanding the definition of likely voters) in NC so I'm wondering if they did likewise in the PA poll. So it may not be so much a shifting of public opinion exactly but a new picture of the expected turnout in the primary.

PPP may be right and their polling adjustment in NC (and possibly this PA poll) may be more accurate now, since turnout will be heavy in both states.

Often, these sorts of discrepancies trace back to sample-weighting. But that's not what's going on here. The two polls (PPP and SUSA) have remarkably similar age, race, and gender profiles.

The major difference seems to be that SUSA pushes voters much harder to make a decision than does PPP - and those undecideds break overwhelmingly for Hillary. So, for example, Hillary leads among white voters (61-33 vs 49-38) and women (62-34 vs 49-39) and trails among male voters (43-50 vs 36-51). In the black community, of course, the dynamic is inverted (75-17 vs 83-17) but that's not enough to significantly dent the overall effect. The trend is clear. PPP shows Obama garnering marginally more support than SUSA, but when the latter survey forces uncommitted voters to voice an opinion, Hillary makes dramatic gains.

In most years, when voter opinions remain highly volatile this far from an election, that'd be good reason to distrust SUSA. This year, though, when voters are following the race remarkably closely - and when undecideds have consistently broken for Hillary in the final days - I'd give the edge to SurveyUSA. So the race is tightening, as Obama wins some converts (particularly, as I've demonstrated before, among Casey Democrats) but Hillary retains a statistically significant lead.

Well the only thing I can say is that he's certainly not losing ground anywhere - he just keeps improving.

Silly PPP. April fools was *yesterday*.

David, could you give some backgrond on Public Policy Polling? A headline on their site reads "Want To Do A Poll?" which does not give me any confidence that they maintain quality control, or any control, over their polling. Perhaps they do everything right, but I would prefer that you provide some background to judge them by and not just slap up a poll because it runs against the grain.

As has been pointed out, this is probably an outlier. BUT, what is more likely:

1. This poll was so poorly coordinated that it's ~15 points off, validating the polls that have Hillary up by around 12, or

2. This poll is something like 7 or 8 points off, which would mean some of the more recent polls are correct and her lead is somewhere around 5.

It seems unlikely to me that a respectable pollster could be off by 15 points or so. This would indicate a relatively small lead for Hillary that Obama could easily overcome by the primary.

Perhaps those who responded to the pollsters were playing April Fools?

It could be that the more Hillary lies, the more people don't like her. ("Obama is taking away your vote!")

I agree with Hopemonger -

With other polls showing the Hillary lead rapidly shrinking, it is not astounding to see one poll with Obama leading by 2 points.

Does this mean I have to start crying again?

Obama has been hitting low 40s in most polls of PA lately. So 45% is not out of the question. It looks like former Clinton supporters might be undecided right now. That is an opportunity for Obama, but they could also come back for Clinton.

I would note two things:

1. The more recent polls show a closer race. Rassmussen was taken 3/31 and PPP was taken 3/31 and 4/1. The Q poll was taken over a long period and therefore does not full reflect BO's swing through PA.
2. I would expect a lot of fluidity in these polls,especially given the number of new registered voters. The variation may be how re pollster are treating these new voters. Note how dramatic the swing was in PPP's NC polling resulting from a cange in their likely voter screen.

I think it's probably more useful to follow the money. Both Democrats are raising more than McCain, but Obama is raising way more than Clinton.

I think the money follows the winner.

The more recent polls show Obama performing better. I think they reflect a bounce after the Casey endorsement and fallout over Bosnia. There are a LOT of veterans in PA and quite frankly, it is an insult to suggest that they would ever allow our first lady and first daughter to fall into harm's way.
I have to say, I'm kind of happy that Billary has been attacking him. He has now proven that he can take a punch. He and supporters; myself included, will pay no attention to the polls and keep campaigning, organizing, canvassing and fundraising as if he were down 10 points. Complacency is for right wingers.

Hmmmmmmmmmmmm.......

Beware of moving goalposts!

If there are people here who actually believe this poll, we play poker every Friday night and if you promise to bring money we will send a limo for you. Just leave a way we can contact you.

And while its unseemly to use this site for wagering purposes and I wouldnt do it,just to make a point, Id would take any wager that, after the primary results are in, this poll is 10pts or more off the final number. When Obama supporters point to a poll like this it shows how desperate they are getting.

The real significance of this poll is not Obama's lead.

It is the UNDECIDED number, which is now up to 12% in this poll.

The Rasmussen poll that came out yesterday showed pretty much the same thing - 11% UNDECIDED.

The Q poll shows 9% undecided

So, if the range in undecideds is 9-12%, that is up from some other recent polls which have shown undecideds as low as 5-6%

I don't think that an increase in the undecided vote means that Obama is doing better. But it does mean that he is getting people to listen to his message & to put their decision on hold until they think about it a little bit more.

I was in New Hampshire when the polls had Obama winning in a blow out. Obviously that didn't nappen. What did happen is that Barack got pretty much what he was projected to get. What made the difference is that almost all of the undecideds swung to Hillary. So, I'm not going to be fooled by the polls a second time.

In focusing on who's leading, too little attention is given to the undecided vote. When the margin of difference, or who's in the lead, is stressed, the underlying assumption is that the undecided sill split evenly between the 2 candidates. Obviously this is not necessarily the case. That's why we have campaigns.

The other factor that's not being considered is the number of new registrations. There were 120,000 new registrations reported last week. Obviously almost all of these people expect to vote, so they represent a higher percent of "likely voters" than they do of the general voting age population. These people would have been screened out by the polls by the question: "Are you a registered voter." They are now included in the polling. It is expected that a very high percent of these new registrations are leading toward Obama & this is probably being reflected in the polling data.

Ahem...

He did do it TX.

GSB27, I don't doubt your viewpoint at all.

My son works in the building Hillary is actually speaking at there in Pittsburgh today and he's been emailing me this morning about it. I was telling him about Lee Hamilton's endorsement and that I thought it would really help Obama in Indiana, and he wrote me this back:

"I get the sense, too, that Obama might do better here than people are expecting."

So you are apparently far from alone in your opinion! I think things are definitely moving Obama's way in Pennsylvania.

Another Super Delegate goes into the Obama column:

Wed Apr 2, 8:50 AM ET

PHILADELPHIA (Reuters) - Democratic presidential candidate Barack Obama on Wednesday won the endorsement of Lee Hamilton, a former Indiana congressman who is a leading U.S. authority on foreign relations and national security.
ADVERTISEMENT

The support of Hamilton, who co-chaired two blue-ribbon commissions that investigated the September 11 attacks and advised President George W. Bush on the war in Iraq, could boost the Illinois senator in his May 6 Indiana primary contest against New York Sen. Hillary Clinton.

Hamilton said Obama offers American voters the best chance to create a new sense of national unity and transcend division.

"He champions the politics of consensus, not of partisan division," the longtime Democratic Party figure said in a statement. "He is driven by the search for the common good."

Hamilton, also backed Obama on foreign relations, an area where the White House hopeful has been criticized for inexperience.

"His foreign policy is pragmatic, visionary and tough," said Hamilton, former Democratic chairman of the U.S. House of Representatives committees on foreign affairs and intelligence.

"He will work with our friends and allies. He will strengthen our ability to use all tools of American power and relentlessly promote the American values of freedom and justice for all people."

It was the latest key Democratic endorsement for Obama in his race against Clinton for the Democratic Party's presidential nomination. The two face hard-fought primary election battles in both Indiana and Pennsylvania in the next several weeks.

Obama also has picked up other endorsements recently from New Mexico Gov. Bill Richardson and Pennsylvania Sen. Bob Casey.

Hamilton represented Indiana in the House from 1965 through 1999. He now is president and director of the Woodrow Wilson International Center for Scholars

Before we dismiss this survey as an outlier, I think it might be useful to read the reported results. Who knows, we might learn something!

For example, I saw that the issue the most respondents cared about was the economy and jobs (48%) And Hillary had a slight advantage with those voters (+2). The second most imortant issue for the respondents was the was in Iraq (29%). With those voters Obama had a slight advantage (+4). Only 8% of the voters said that healthcare was the most important issue. That is Hillary's big issue and she did the best with those voters. But not as many people care about it as much as she does in Pennsylvania.

I also noticed that there are some people in Pennsylvania who are neither black nor white. The survey refers to them as Other. And those others went 50% for Obama, 25% for Clinton. I think that is potentially significant going forward.

From looking at the other poll questions it appears to me as though Pennsylvanians are going to vote according to their best interests economically above other considerations. And who can blame them? But the war is still a significant issue and one that both candidates need to address in MEANINGFUL ways in the weeks to come.

This is likely an outlier, however, it shows movement and right now all that is going to Obama. He needs to keep the pressure on here really force HRC to devote all her resources to PA and if he can lose by 5% or less then job done. BTW did I jsut see Wyoming Gov endorse BHO?

Man, I knew this one would have the inmates banging their cups against the cage. Although, actually, it's been calmer than I expected.

But stop dissing PPP. It has a damn good record in this election (and they're not too proud to tell you about it: http://www.publicpolicypolling.com/pdf/ppptrackrecord.pdf.)

And if we can all stop screaming “outlier” for a minute and engage in some actual analysis, a comparison of the methodology, questions and the crosstabs in this one to the Q-Poll will show that they aren’t really all that inconsistent.

First, PPP is a robopoll that gives the respondent three choices, Hillary, Obama or undecided, and then moves on. If you’re even a little ambivalent, there’s no pressure to choose. Q-Poll, however, has an actual underpaid college student at the other end of the line who is required to lean on the undecideds to try to squeeze out a “lean toward” position. These leaners are included in Q-Poll’s results (and I didn’t see any date back them out).

Second, consider the dates and the sample size. PPP was taken 3/31 to 4-1, n=1224. Q-Poll was taken 3/24 – 3/31, n=1,549. PPP is very recent and very compressed, compared to Q-Poll. In a slow news week, Q-Poll’s leisurely pace wouldn’t be a problem and might have some benefits.

Does anyone think 3/24 – 3/31 was a slow news week for those following the Democratic primary?

Over the course of that week, a) Obama’s race speech was breaking records at Youtube and the Wright flap was dying out as all but the hardcore Hillaristas and the Fox-watching wingnuts deemed themselves satisfied, b) Tuzlagate kept getting worse for Hillary, so she sat down with Richard Sciafe to make a graceless, hamfisted effort to revive the Wright issue, c) the Red Skull called Richardson Judas on Easter Sunday and then unapologetically did it again on the OpEd page of the WaPo, which offended many devout Christians and reminded many who are not committed Clintonistas of the parts of the 90s they didn’t like, d) the MSM suddenly started paying a lot of attention to the math, and, e) as has happened in so many other elections this cycle, the very definition of “likely voter” changed as the Obama (and Clinton) voter registration drives ended.

A poll taken 3/31-4/1 would tend to capture the end result of all that stuff. One done over the course 3/24- 3/31, to say the least, would not.

Third: compare the toplines:

PPP(3/31 to 4/1, 1,224.): H 43% , O 45%, Und. 13%
Q-Poll (3/24 – 3-31 1,549 “likely Democratic voters”): H 50%, O 41%, Und. 8%, Other 1%

Between the way Q-Poll pushes leaners to choose (which PPP doesn’t), the timeframes and the 5% difference in the “undecided” numbers, it’s hard not to speculate that Hillary’s Q-Poll number includes a lot of soft support reacting to the Worst of Wright reel who drifted back to Obama or “Undecided” by Monday of this week. It’s also interesting to note that Obama’s numbers in the Q-Poll and the PPP poll aren’t that different. Indeed, as an Obama supporter, that 13% is cause for concern. We know how late undecideds have been breaking in most of these rustbelt states. If they aren’t in his column by the day before the election, Hillary gets ‘em. The problem for him is that the demographics would tend to indicate that’s even more likely to be true in Pa., and there are a lot of them right now.

Finally, look at the crosstabs of the two polls; they look an awful lot alike. That tends to indicate that these two polls are not as radically inconsistent as many are saying. And there’s data to encourage and concern both camps within them.

On the upside for Obama the crosstabs show much the same pattern we've seen in almost all of the states Obama has won: he dominates among African Americans, ties among men, leads to a greater or lesser extent in every age group except for the 65+ group and loses big among white women and people over 65.

On the upside for Hillary, Pennsylvania seems to have more older white women and people over 65 than your average state and that 13% total undecided is disproportionately white or “other (which I’d guess is mostly Hispanic), and over 65. That’s EXCELLENT . . . well, you know.

look at the movement of that PPP poll and you'll see just how ridiculous it is. But if it gives you obamamites HOPE, well then

KEEP HOPE ALLLIVE!

ROFLMAO

LOL LOL LOL

hee hee heeee

HHAAAA HAAAA HAAAAA !

IEEEEIIIIEEEEhah!

Even if this poll IS an outlier (PPP has a good track record) this is absolutely horrible news for Clinton. Wasn't she ahead by 20 at one point?

Combine this with the news that her campaign is bouncing checks like ping pong balls and you can see why her endorsements have dried up.

According to this Gallup Poll, Obama is ahead of Hillary by a few points and has not dropped more than 3 points. Vlinton is down 2 points from where she was two weeks ago.

http://www.gallup.com/poll/105970/Gallup-Daily-Obama-49-Clinton-46.aspx

I know that elected Democrats may be going towards Obama because they know that we need change in this nation. I think as Americans we feel shame in what our government has done.

I think this election is America's chance to redeem itself and clean house and get on the right path. Those in power have taken advantage of their positions and dragged our nation through the mud, as well as Iraq.

I think true Democrats and Republicans even realize their failure in approving the war and the consequences of that.

I think they are redeeming themselves by going with Obama, because we are sooo far off the path of morality, integrity and democracy. I think they are going to stop this nation from slipping further. There is a line, and it was crossed long ago, no?

As an Obama supporter, I'd love this to be true, but the poll seems is a little unlikely. I very much doubt Obama wins PA outright.

BUT Obama doesn't NEED to win PA. He needs to prevent Hillary from scoring a blowout. All the polls over the past few days (not just this possible outlier) suggest he may well be able to do just that, and then will completely erase any delegate gains she might have made when NC votes for him by a 20% margin in early May.

PPP's polling question:

The Democratic candidates still running for President are that stupidhead Hillary Clinton and everyone favorite candidate Barack Obama. If the primary was today, who would you vote for? If you would vote for the lying spoilsport Hillary Clinton, press pi calculated to 27th decimal. If for Barack Obama, press 2. If you’re undecided, that's OK. Just press 2.

(Just screwing around, obviously. I don't know what the heck to make of those results.)

I also tend to discount a 28 point swing over such a short period, but it is possible that a combination of the Casey endorsement and the utter bizarreness of her meeting with Richard Scaife has moved her numbers down somewhat.