Prolonging cruel game, Pennsylvania voters satisfy neither candidate

Once again, neither candidate blinked. In this trench warfare, Hillary Clinton and Barack Obama held their ground, producing a result that satisfies no one and only strengthens the status-quo. Given a third opportunity to send Clinton packing after New Hampshire and the March 4th contests, Pennsylvania Democrats chose to prolong the race and they did so decisively, leaving no room for Clinton to drop out even if she wanted to. Yet, just as had happened the two previous times, enough voters rallied around Clinton to make her victory credible but not enough to improve her chances of seizing the nomination.

The Democratic electorate seems to be thinking of these primaries as a cruel game through which it can torture its candidates -- making sure neither gets a result that could meaningfully help them. As a result, the race is showing no sign of slowing down and the candidates are already on to May 6th, the next mini-Super Tuesday featuring North Carolina and Indiana.

In my preview of the race yesterday afternoon, I differentiated between three issues: first, what Clinton needs to remain in the race; second, the threshold of a credible victory; third, what would allow her to change the fundamentals of the race. There were plenty of possible results that could have left room for spin, but last night's results contained very little ambiguity:

1 - By posting a comfortable victory, Clinton undoubtedly gained the right to stay in the race and considerably eased the pressure to drop out. After all, how can a candidate who just won such a large victory in a crucial and hard-fought state even think of dropping out? Note that one major problem for the Clinton campaign is money; they are not paying bills and they have almost no money left in the bank. If the financial situation does not improve quickly, Clinton could be forced out of the race because of lack of funds. She was counting on tonight's results to energize her base and get money flowing back in and for now she seems to be succeeding: She raised a stunning $2.5 million in the four hours after the polls closed last night.

2 - Despite falling just short of the double-digit mark, she passed the threshold of a credible victory, one whose legitimacy even the Obama campaign will not attempt to question; she barely reached this level, to be sure, but the tone of the media coverage and of this morning's headlines leave no doubt that Clinton met the expectations that had been set for her. "With Clear Victory, She Has Rationale to Fight on," writes the New York Times today. Keep in mind that the Obama camp was fully prepared to spin a narrower contest as a moral victory.

3- Just as evident, however, is that Clinton's chances to seize the nomination did not improve; if anything, her prospects are even more somber this morning than they were yesterday. Pennsylvania was one of Clinton's last shot at turning the race on its head, at cutting into Obama's delegate total or at showing that Democratic voters are increasingly turning back to her after a flirtation period with Obama. She did none of these things: While the exact delegate breakdown is still unclear, she needed twice as big a victory only to leave it possible that Obama not emerge out of the primaries with a large pledged delegate lead; after yesterday's vote, Obama is virtually assured that he will get to June with a significant advantage among pledged delegates.

On a more symbolic level, Clinton could have demonstrated that something has truly changed in the Democratic primary had she won by 15-20%. At the beginning of the Pennsylvania campaign, such a margin seemed to be very much possible, with some polls showing Clinton expanding to the 20% range during the Wright controversy. Just imagine how damaged Obama would have looked tonight had Clinton pulled such margin off.

Backed into the corner, Clinton undoubtedly survived to fight another day; but she did not move the numbers. In fact, the extent to which the numbers have held remarkably static since Ohio is truly remarkable: Not only is Clinton's lead the same, but the voting pattern of most groups is similar, with Hillary's winning margin slightly decreasing among white men and slightly increasing among Catholics. In other words, Clinton did not demonstrate that those weaknesses have increased over the past few weeks as she was hoping to do; after all, the Wright controversy and bittergate were supposed to have hurt Obama among these voters.

Naturally, none of this is to deny that the inability to move number is as much if not more Obama's failure than Clinton's. Pennsylvania once again served as evidence that Obama had fundamental problems relating to blue-collar voters: The Illinois Senator got trounced in most of the state's working-class or rural counties, and was stuck in the 20s in a number of them -- a stunningly poor showing. Even worse, Obama did not bring about strong turnout among young voters and his own position among white-collar voters was much weaker than usual yesterday; he for example barely held on to voters with a college degree. And all of this despite the fact that he massively outspend her and has been campaigning as the inevitable nominee for quite some time now. If he wins the nomination, Obama will have to urgently address his significant weakness among blue-collar voters if he does not want the Reagan Democrats to desert the party once again.

But this is no longer New Hampshire, nor is it even Ohio or Texas. Pennsylvania was not a zero-sum game but one of the last primaries in a long series of contests that started in early January. Since then, both candidates have held together very solid electoral coalitions but Clinton blinked a few times too many, leaving her trailing in most important counts. Her most important audience now is superdelegates, and she has been somewhat successfully making her case that Obama would go in the general election with glaring weaknesses but that has not proven enough to move many superdelegates her way. A 10% lead in Pennsylvania coupled with the exit polls we saw yesterday help her make her case, but it is only enough to stall for more time not to generate movement towards her.

Now, the campaign moves on to further contests. As always, the question will be whether either candidate can transcend the demographic logic that has determined almost every one of these Democratic primaries. Until he finds a way to do so and however inevitable his nomination looks, Obama will not be able to put Clinton away and both candidates will be forced to go through the motions of a competitive race. Given how much Obama prides himself for his ability to bridge the country's stubborn divides, it is ironic that it is his failure to make inroads in his opponent's demographic base that is dragging this primary longer than anyone thought possible.

Pennsylvania results thread: Clinton prevails; Childers 400 votes from stunning pick-up

12:45am: With 98% reporting, Clinton's lead is back down under 10% -- at 9.4% to be exact. Most of the remaining precincts look to be in Chester County in which Obama is ahead, so it does not look like Clinton crossed 10%. However, her win clearly falls within the "credible victory" category I outlined this afternoon -- though it is certainly not enough to change the fundamentals of the race. Clinton had the potential to score a much larger victory, and she reach the level she needed to truly change the discussion. As for delegates, Clinton failed to meaningfully dent into Obama's margin, and this was by far the largest state left, ensuring that the primaries will end with Obama holding on to a very large lead among pledged delegates.

Yet, how can a candidate who just won such a large victory in a crucial state even think of dropping out? Clinton has to be happy with today's narrative since she won based on her strength in rural areas and blue-collar voters. She broke 70% in many counties in Western Pennsylvania (79% in Fayette County). And in the all-important issue of money, it looks like Clinton will be able to generate money out of tonight's victory. Her campaign claims that she had raised $2,5 million from the polls closing as of 11:30pm! More on all of this tomorrow. And also on MS-01, where the last county got Childers even closer to 50% -- about 400 votes short. But this race is on to May 13th.

11:20pm: Clinton is holding firm to her double-digit lead with 88% reporting. In short, the results today are remarkably similar to those of Ohio, with most numbers internal numbers today very similar to those we saw on March 4th. Most things have remained stable in the past 6 weeks: Obama moved numbers very little despite massively outspending Clinton, and Clinton did not move upward among blue-collar voters despite her argument that Wright and bittergate hurt Obama in that group.

10:50pm: The AP has called a runoff in MS-01. With only 1 precinct remaining, Democrats just missed a pick-up in a conservative Southern district. Childers has 49% of the votes versus 46% for Davis but he remains only about 470 votes from the 50% threshold! This will certainly get Childers and the DCCC very frustrated, but it also guarantees that there will be a lot of spending in this race in the weeks ahead.

Meanwhile, Barack Obama has taken the stage in Indiana, already looking ahead to the now crucial contest of May 6th. He tried to put Pennsylvania behind him, using a tactic Clinton employed in late February when she gave her speeches on nights of defeat from the upcoming state. After congratulating Clinton for her victory (something she did not always do), Obamalaunched into a long charge against John McCain. He then moved on to his attacks against Clinton, drawing many contrasts without mentioning his opponent's name; he called to reject a party that uses the fear of terrorism to win elections (a clear reference to Clinton's 3am and bin Laden ads), tests its message in polls and relies on divisiveness. Meanwhile, Clinton has progressed to a 10% lead with 79% reporting.

10:40pm: In MS-01, Childers is now ahead by 500 votes... but both candidates are now far away from 50%. Results can be found here. The only outstanding county right now is Clay county, which Kerry won with 52% -- so it is very likely that Childers will stay ahead but it is very unlikely that he gets above 50%.

In the Democratic primary, meanwhile, 76% of the votes are in and whether Clinton can inch to a 10% victory will determine much of the coverage of the race in the days ahead. As I explained this afternoon, she needed a double-digits victory, or at least a high single-digit lead to have what I called a "credible victory." But she has certainly not reached the level she would need to change the fundamentals of the race.

10:20pm: Speaking in Philadelphia, Clinton vows to press on, emphasizing that, after a long campaign in which both candidates criss-crossed the state, Pennsylvanians chose her . "The American people don't quit, and they deserve a President who doesn't quit either," she said. She also insisted on the need for people to go to her website and donate... Her campaign is in dire need of money, though they are reporting that they have raised half a million dollar tonight already. With 68% reporting, Clinton is still in single-digits. Both Philadelphia and Pittsburgh have almost finished reporting.

Meanwhile, exit polls have been updated. Among notable changes: Clinton won big among women (57%) but the surprise is among male voters: Obama only held on to a 4% lead. Helping Clinton, of course, is the fact that 59% of voters are women. Clinton is now ahead 62% to 38% among white voters and broke into double-digits among African-Americans (not that that is a strong showing at all); she is winning big among Catholics (69%) but she is also ahead among Protestants, contrary to what some polls were finding. Clinton wins union households with 59% and voters with no college degree voted for her with 58%; Obama underperformed among voters with a college degree (51%). And in a very important measure, Clinton is found to have won the white vote 56% to 44% -- that is still less than her margin in Ohio, an argument Obama will use when talking to superdelegates.

10:10pm: With 92% of precincts is reporting, it looks like the MS-01 is heading for a runoff with Davis leading Childers by 400 votes, 48% to 47%. More proof that Childers is benefiting from strong turnout from Democratic voters: In Itawamba County, which voted for Bush overwhelmingly in 2004, the Democrats won with 59% today. There are plenty of similar examples.

Some interesting notes on the presidential race: Ben Smith finds that, among Republicans who went to vote in the meaningless GOP primary today, ie. the base of the base of the GOP, only 73% say they will vote for McCain in the general election suggesting that McCain still has some work to do among conservatives. Many of those could be Ron Paul supporters, as the Texas representative is coming in with 16% right now, ahead of Huckabee.

10pm: Quite a suspense in MS-01; with 83% of precincts reporting Davis is now at 49%, barely under the 50% threshold and 1,100 votes ahead of Childers. 4 counties have significant number of votes to count; I just looked back at the 2004 elections and one of these counties voted for Kerry, so Childers will score big there. Of the three others, one voted for Bush at the level of the rest of the district (61%) and two are more Republican than the district at large. On paper this favors Davis but don't forget Childers is significantly overperforming the district's partisan make-up.

In Pennsylvania, half of the precincts are now reporting and Clinton is ahead by 8% still; Pittsburgh and Philadelphia are now reporting at the same level, way ahead of the rest of the state. That suggest that most of the outstanding precincts are from non-urban areas, which should favor Clinton. A CNN commentator is suggesting that a Clinton nomination would make many voters flee the party -- though more Clinton supporters in Pennsylvania are saying they would vote for McCain if Obama is the nominee than Obama supporters.

9:50pm: Davis is now ahead 51% to 45% with 71% of precincts reporting. Davis's stronghold, DeSoto County, is now entirely in; Davis won there with 81% (8000 votes), which is 400% of his total margin. There is thus the potential for Childers to hold the Republican under 50%.

In Pennsylvania, Clinton is ahead by 8% with 45% of precincts reporting. Obama is performing better in Philadelphia (61% to 39%) but 72% of the city's precincts are already reporting. Clinton is swamping Obama outside of Philadelphia, posting very strong numbers in most of the counties heavy with blue-collar voters, with way above 66% in many counties. It is too early to talk about delegates, but this type of margin will enable Clinton to get strong results in many even-delegate districts.

9:25pm: Clinton's margin of victory is still unclear -- Clinton is leading by 6% with 15% of precincts reporting. It is too early for the Clinton campaign to be triumphant -- though there are indications that her lead might grow significantly because of the simple fact that 41% of Philadelphia is reporting. PA has been called for Clinton, but the race's dynamic going out of this contest is not yet determined.

In Mississippi, there has been a change as counties in which Davis is strong just reported. He is now ahead 50% to 46% with 46% of precincts reporting. If Davis wins tonight, the DCCC will regret not having met the NRCC's spending and waited for the May 13th runoff.

9:10pm: The danger for Obama is that the storyline is that Clinton's win is due to blue collar voters. With 10% reporting, Clinton is ahead by double-digits and Philadelphia (Obama's stronghold) is reporting 28% while Pittsburgh (where Clinton is strong) has not reported at all.
In Mississippi, with 25% reporting, Childers is ahead 53% to 43% -- hanging on above 50%.

9pm: AP calls Pennsylvania primary for Hillary Clinton.
It looks like Clinton's actual winning margins are bigger than those in the exit polls -- which also suggests that the numbers among key groups like white men and blue-collar voters will likely change to suggest larger wins by Clinton. Other good news for Clinton: I said that her winning margin among white men was cut since Ohio, but it looks like it has increased among other key groups like Catholics or union households; this will be used to push the argument of Obama's weakening. With 25% reporting in Philadelphia, Obama is only getting 55% -- a decent margin, but much less than what he would need.

8:45pm: Raw votes are starting to come in, and with 3% of precincts reporting Clinton is ahead 55% to 45%. This includes 11% of Philadelphia -- in which the two candidates are tied right now! Exit polls suggest Obama got a huge victory in Philadelphia, so this implies the most favorable neighborhoods have not reported yet at all (or so Obama should hope).

In Mississippi, there are some very important numbers: With 7% reporting, the Democrat, Childers, is leading 58% to 37% against Davis. It is unclear where those votes are coming from, so things can obviously get much tighter, but the truly significant result for now is that the other candidates are getting 2% at most. This is a surprise considering that the ballot was non-partisan and that the primaries had been very nasty. This means that the odds of a candidate crossing 50% is much higher than previously thought -- we might have a winner tonight.

8:25pm: The all important white male vote will be scrutinized very closely in the coming days, and the exit polls as they stand right now are not bringing her good news: She won the white men vote 53% to 46% (she won the white female vote 64% to 36%); in Ohio, she had won among white men by 19%. One of Clinton's main hopes was to claim that Wright and bitter-gate weakened Obama among white voters; it's going to be hard for her to argue that Obama has slipped dramatically enough to endanger his general election position if these numbers are confirmed in later updates to exit polls. However, Clinton won Catholics with 68% -- another group that we were looking for closely.

8pm: Polls are now closed in both Pennsylvania's primary and MS-01. As expected, there is no call in the Democratic contest; this does not mean that the race is going to end up close. It took a while for Ohio to be called and Clinton won by double-digits. Wolf Blitzer claims that "If Clinton wins tonight she will of course go on," an interesting expectation setting with which the Obama campaign would beg to differ. By the way, Clinton is in Philadelphia, Obama is in Indiana.

The full exit poll just posted on CNN's website shows Clinton narrowly ahead by 4-5%. This now looks to be the late wave of exit polls -- which tend to be much less skewed than the second wave. Among interesting numbers: 14% of voters changed their registration to Democratic since January, and 60% of them favored Obama (So much for those who criticize Clinton for riding the Rush Limbaugh Operation Chaos wave). Clinton leads by 14% among voters with no college degree; Obama leads by 8% among voters with a college degree. These numbers will be refined through the evening.

7pm: In Pennsylvania, a group sought to keep the polls open in Philadelphia until 10pm but the motion was denied, confirming that the polls will close (and results will start tricking in) at 8pm. In Mississippi, I reported that polls had closed but, as a commenter correctly points out, they close at 7pmCT -- so not for another hour.

Original post: The polls close at 7pm in Mississippi and at 8pm in Pennsylvania. Most of the suspense, of course, is in the latter. The first exit polls are starting to leak -- and as always these numbers should be taken with a huge grain of salt since they include only part of the day. And please keep in mind that this is only the second wave of exit polls and those have often been too favorable to Obama: The ones from Ohio showed a tie on March 4th, and Obama was found to be leading in Massachusetts on February 5th (he lost both states by double-digits). In fact, Pollster.com's Blumenthal finds that 18 out of 20 second wave exit polls were too skewed to Obama's favor by an average of 7%.

That said, Fox reports that these early exits have Clinton leading 58% to 42% among gun owners, with Obama ahead 54% to 46% among voters with a college degree -- perhaps a smaller margin than he would need. CNN adds that these exits are finding Clinton is ahead 55% to 45% among white men, that Obama getting as massive a victory among black voters as ever (92-8) and that those who have decided in the past week heavily broke for Clinton. As for full results, there are being leaked indirectly, they don't agree with each other and thus don't look reliable. The National Review (which often gets correct exit polls first) is reporting that exit polls have Obama narrowly ahead. Drudge is showing a narrow Clinton lead but Marc Ambinder notes those are the first wave of exit polls -- not even the second.

For now, you can read my guidelines of what to expect tonight here.

Setting tonight's guidelines: Will Pennsylvania's primary be tighter than Mississippi's special?

This is the third Election Day in which Barack Obama has been in a position to bury Hillary Clinton; he came short on January 8th in New Hampshire and failed to do so again in Ohio and Texas on March 4th. From now on, of course, every Election Day could mean instant death for the Clinton campaign who has no room for error all the way until June 3rd. To continue using the metaphor of a tennis match, Obama has quadruple match points -- and some of them (North Carolina, Oregon) are on his serve.

Clinton is hoping that a solid showing today in Pennsylvania will help her finally get some momentum and head into the May contests in a strong position. Polls over the past few days suggest that Clinton is heavily favored to win, but the margin of victory is more unclear; few polls are showing anything more than a 10% lead and the key to tonight's results seems to reside in the turnout of groups that typically favor Clinton (blue-collar whites, Catholics).

The issue, of course, is defining what a strong showing would consist in. There are, as was the case in Ohio and Texas, three separates issues to consider: (1) What result does she need to stay in the race? (2) What result would make her claims of victory credible? (3) What result would change the fundamental dynamics of the race? It's worth addressing them in turn.

(1) What Clinton needs to simply stay in the race is perhaps the hardest question to answer because it fully depends on Clinton's determination; only she can decide to bring an end to the race by dropping out. It would be very difficult for her to stay in if she loses Pennsylvania; but what about a narrow victory (say 3-4)? That would be interpreted as a very weak result for the Clinton campaign; but would she pull the plug?

(2) The threshold of a credible victory is fairly clearly defined: Clinton needs a double-digit victory, or at the very least a very high single-digit lead. Under this "Ohio threshold," the Obama campaign will claim to have exceeded expectations and pride itself on a moral victory. Above it, it will be nearly impossible for Obama to spin this into a decent showing and questions will only build as to why Obama is unable to close this off.

(3) To change the fundamentals of the race, Clinton needs much more than a double-digit victory. Her surprising comeback in Ohio, after all, allowed her to stay in the race but it did not alter the primary's basic dynamics. Backed into a corner, Hillary needs a dramatic showing to get some air -- and that involves exceeding expectations with a comfortable double-digit victory of at least 15% and she also needs exit polls to show Obama has come in very weakly among white blue-collar voters. Even then, of course, Clinton would remain the heavy underdog but she would finally have succeeded in moving the race meaningfully.

These measures might seem unfair, of course, considering that Obama has massively outspent Clinton; not to mention that, considering how close she is from the exit, it is remarkable that Clinton is still standing -- let alone in a position to win a primary as important as Pennsylvania's. But this year's contests have been defined by demography: Both Democratic candidates have their core constituencies, and neither has made significant inroads in the other's base groups. If anything, African-Americans are more determined to vote for Obama than they were in January and downscale whites are looking more solidly in Clinton's camps than they were at the end of February.

In this trenchware warfare, the slightest weakness in either candidate's core groups have cost the candidates greatly (Clinton in Wisconsin) but unusual displays of strengths have yielded great rewards (Clinton in NH and OH, Obama in SC). If Clinton blinks again, it could prove fatal to her campaign; but if she keeps her groups mobilized today it could be very damaging to the Obama campaign.

Mississippi's 1st district: Also today, voters in the MS-01 are going to the polls for the first round of the special election to replace Rep. Wicker. This has become a surprisingly contentious election considering how red the district is; the NRCC have dumped almost $300,000, more than twice the amount spent by the DCCC, testifying to how panicked the GOP has become at the prospect of unexpectedly losing a conservative Southern seat.

The two main candidates are Democrat Childers and Republican Davis -- but the ballot will feature 6 candidates total, including two important politicians (one from each party) who lost to Childers and Davis last month in the primary for the actual November election. They are not campaigning for the special election, but their presence on the ballot makes it almost impossible for either of the main candidates to get elected tonight.

If no one reaches 50% today, a runoff will be held on May 13th; but tonight's results are certainly very important for they will tell us how strong a base vote Democrats have and how much bitterness there still is among Republicans after a very contentious primary. Depending on the answer to these questions, the DCCC might decide to invest more in the district or to give up the fight -- and that will go a long way toward determining what will happen in the runoff.

Tuesday polls: Should we paint New York purple?

The six-week lull is over, and it's Election Day! The afternoon and evening coverage will be heavy with Pennsylvania analysis, of course, so let us start with the day's polls -- many of which are not Pennsylvania related.

• Zogby released his last tracking poll from Pennsylvania, finding Hillary Clinton jumping to a 10% lead, 51% to 41%. There is a clear trendline in favor of Hillary in the final days, as her lead moves from three to six to ten points.
• Zogby confirms which group holds the key to today's election: white men, particularly Catholics.

• Insider Advantage, meanwhile, also released its second poll in two days and finds Clinton's lead at 7%, 49% to 42%. Yesterday's survey showed a 10% lead.
  

The expectations have been set by the wave of polls released yesterday, all but one showing Clinton in the lead with anywhere from low single-digits to low double-digits. If Clinton stays in that range, this is likely to be a disappointing night for her; if she goes beyond it and wins by solid double-digits, we are in this race for the long haul.

Naturally, as early as tomorrow morning we will be on to the May 6th primaries and things are not going very well for Hillary there either, with the North Carolina contest increasingly looking like it could sink her:

• PPP's latest poll released yesterday shows Obama opening a 25% lead against Clinton, 57% to 32%.
  
• Note that PPP is the only institute that has shown Obama leading in Pennsylvania -- and has done so consistently. Depending on today's results, we will know if their turnout model is good or if it has flaws.

There is no doubt that Clinton cannot survive May 6th if she suffers such a loss, whatever happens in Indiana. First, it would only deepen her delegate predicament; second, it would be a terrible symbolic blow in one of the country's biggest states; third, such a large Obama victory would undoubtedly be accompanied by his progressing among the groups Clinton wants to show he is weak in. Naturally, if Hillary posts a very strong victory today it could help her down the line (not that we have seen momentum carry the candidates to further victories this cycle).

Finally, a wave of general election polls:

• In New York, first, Siena University paints a very worrisome picture for Democrats as McCain is extremely competitive against both Hillary Clinton (46% to 42%) and Barack Obama (45% to 40%). The favorability ratings are also remarkable, with Clinton only standing at 48-46 in her home state, behind both McCain (54-35) and Obama (54-34).

• Rasmussen confirms that Democrats have no chance in Arizona, McCain's home state, despite its having been a swing state in past elections. The Republican beats Clinton 60% to 32% and Obama 57% to 37%. Note that there are competitive House races in Arizona (AZ-01 and possibly AZ-05 and AZ-08) in which McCain's strength could help the GOPers.

• Finally, a national poll by USA Today/Gallup (this is not the tracking poll) shows both Democrats posting a narrow lead over McCain. Clinton leads 50% to 44% and Obama is ahead 47% to 44%.

There have been enough New York polls showing a tight race at this point that Democrats should start getting worried; the Marist poll two weeks ago showed Obama actually trailing McCain; both Quinnipiac and SUSA's last week showed Obama held in single-digits (though Clinton was in double-digits). Despite the Siena poll's assertion that Clinton is as weak as Obama in a match-up, the Illinois Senator typically has more trouble in past polls.

It would be a catastrophe for Democrats to have to spend any money or time in one of the country's most expensive media markets in order to defend one of the states that has been the most reliably Democratic in past cycles. In fact, the New York Republican Party is slowly but surely dying -- but it would be a great blessing for them if McCain's coattails were unexpectedly strong enough to save their majority in the state Senate.

It is still unlikely that New York becomes competitive -- but we will have to keep an eye on the polls. After all, at the heart of McCain's general election strength is his appeal to independents and moderate Democrats.

In battle for blue-collar vote, guns emerge as hot issue

The two Democratic candidates are trading harsh attacks on the issue of guns -- a topic that has been escalating in this campaign ever since the coverage of Obama's "bitter" comments started. But, somewhat paradoxically since this is a Democratic primary, they are both trying to display their anti-gun control credentials.

The Obama campaign has been contacting Pennsylvania voters with robocalls blaming Clinton for her anti-gun record, in particular her vote "banning the confiscation of weapons in emergencies." As the ad claims, "Barack Obama respects our traditions. Senator Clinton voted to allow guns to be confiscated, and this raises real doubts."

Clinton's response is as brutal, as she launched her own robocalls charges Obama with "not telling the truth about his past position on the Second Amendment:"

As an Illinois state Senator, he supported a ban on all handguns, and he even personally filled out a questionnaire saying he supported the ban... And Obama voted to give millions of dollars to anti-gun organizations that are trying to take away our gun rights.

This is just another example of Barack Obama saying one thing, and doing another -- and yet another issue John McCain will use against Obama to win in November.

The vigor of this gun exchange and both candidates' concern with proving their pro-gun credentials shows how essential the blue collar vote has become in tomorrow's vote. This wasn't necessarily the case in all the previous primaries; in many contests, Obama was targeting other groups that were close to Clinton. For a long time, Obama targeted African-Americans, a group he successfully rallied to his side at the beginning of 2008; subsequently, he focused on the female and Hispanic votes which he failed to significantly move in many high-profile contests.

Yet, it is Clinton's strength among blue-collar voters that has emerged as the most significant danger to Obama's nomination as this is Clinton's main argument for why Obama is unelectable (or less electable) in November. And the controversy over Obama's bitter remarks were seized by the Clinton campaign precisely to drive Obama's numbers down among this group of Pennsylvania voters; Clinton is hoping that, after tomorrow's vote, she will have more evidence that Obama has a potentially fatal weakness.

Obama knows he does not necessarily need to win Pennsylvania, nor does he need to prevail among downscale voters; but he does want to cut Clinton's margins to undermine her electability argument and the gun argument is one tactic his campaign has found to attract some blue-collar votes, or at least to drive down their enthusiasm for Clinton and get them to stay at home. Remember, this is the group whose turnout will likely determine tomorrow's result.

In brief, we will parse through exit polls as soon as they are available tomorrow for any evidence of the breakdown among lower-middle class voters, union households, etc. More than the actual results, these are the numbers that could drive the conversation in the days (and weeks?) ahead.

Clinton closes Pennsylvania campaign with negative spot

In the closing days of the Ohio and Texas primaries, the Clinton campaign had come out with the first version of the 3am phone call ad -- as direct an attack on Obama's national security readiness as any Hillary had attempted.

Seven weeks later, the Clinton campaign once again introduced an attack ad questioning Obama's experience, one that is arguably even more controversial than the previous one (You can watch it here). "It's the toughest job in the world," says a voice as the clip goes through a series of images, evoking the stock market crash, Pearl Harbor, the Cuban missile crisis, Bin Laden, and Hurricane Katrina. The ad concludes: "You need to be ready for anything -- especially now, with two wars, oil prices skyrocketing and an economy in crisis. Harry Truman said it best -- if you can't stand the heat, get out of the kitchen. Who do you think has what it takes?"

The ad is bound to be controversial in Democratic circles because it uses a shot of Bin Laden, perhaps the tactic that is the most characteristic of the "politics of fear." It furthers the argument made in the 3am and it extends it from a national security worry to a more general attack on Obama's inexperience, as many of the issues this ad touches are economical.

Many Democrats will legitimately worry that this is a theme McCain will eagerly seize upon and drive much further in the general election if he faces Obama. In fact, the series of catastrophes the ad races through are reminiscent of some of the ads Rudy Giuliani ran last fall; Giuliani was fond of paranoid ads listing all the threats and enemies facing America. This points to a fundamental weakness in Clinton's ad: this ad only does a tenth of what Giuliani's ad dared do (and probably what McCain's will do) -- and, naturally, Clinton cannot strengthen it any further without really attracting the ire of her fellow Democrats.

Can this kind of attack ad be truly effective if the script cannot be as strongly worded as it would need to be to have an impact? The same question applied to the 3am ad, despite its being credited with moving numbers on March 4th. I think the controversy over Obama's position on NAFTA was much more damaging to his candidacy than the 3am ad, but if the 3am did have an impact, this latest ad is weaker than its predecessor in that it makes its national security point much more indirectly.

Yet, this latest ad is stronger than the 3am ad in a decisive respect: It is centered on domestic issues. This is Clinton's strong suit -- not foreign affairs. Obama has had a ready response to Clinton's claims that she is more experienced when it comes to national security (the Iraq vote) and the recent controversy over Bosnia undermined Clinton's reputation on foreign affairs even further. But voters still relate to Hillary on the economy and she is widely credited for her mastery of policy. Another reason why negative ads on domestic issues are stronger than those on national security: Clinton cannot be accused of hurting Obama's cause in the general election, for McCain would have much more difficulty using a readiness argument centered on the economy than one centered on national security given McCain's own economic inexperience.

With that in mind, it is puzzling that Clinton has not more forcefully insisted on the argument that she is more ready to address these kinds of economic emergencies (foreclosures, health care, oil prices) -- rather than go back to 3am national security phone calls and spots about Al Qaeda.

The shot of Bin Laden makes the ad more controversial while, in my opinion, diluting its impact, so why include it? Clinton's bet is probably that it will help her earn free media coverage. The ad was launched today, in the final hours of the campaign, and unless the ad gets picked up in evening news and in cable shows it is unlikely that a sufficient number of voters would see it enough to think about it before voting.

It's all about turnout: Monday polls fail to set expectations

A wave of new polls released today should set the expectations for tomorrow's primary, but they instead confirm that the results will remain suspenseful to the end, as there is very little constitency across polls, neither in trendlines or in numbers:

• SUSA's final poll, first, shows a tightening of the race, with Clinton's lead to a 50% to 44% margin. Last week, Clinton was leading by 14%.
  
• The internal numbers show nothing surprising: Clinton is very strong among women (+23%), whites, and voters with no college degree (+15%).

• Suffolk University released its first Pennsylvania poll and finds a 10% lead for Clinton, 52% to 42%.
  
• An interesting internal number is that, of those who watched the ABC debate, 46% were more impressed by Clinton and 26% by Obama.

• Quinnipiac's final poll has Clinton gaining a point from last week and leading 51% to 40%. Quinnipiac confirms SUSA's finding that the key battleground tomorrow will be Southeast Pennsylvania.

• Rasmussen's final poll shows a tightening of the race, with Clinton leading 49% to 44% -- a 4% drop since the end of last week.

• Insider Advantage confirms that a double-digit victory is within Clinton's grasp, showing her up 49% to 39% -- with 12% undecided.

• Finally, PPP complicates things by finding Obama leading, 49% to 46%. This is the same margin that had been found last week.
• PPP is the only institute to have ever found Obama ahead in Pennsylvania -- but they have done so three times now. Their turnout model clearly is one that projects a lower turnout among whites and blue-collar voters.

The situation is closer to the one in Iowa than perhaps any other election since January 3rd. In other words, Pennsylvania will be decided by which groups turn out. From Iowa onward, both candidates tried to make inroads in the other's core constituencies, with Obama courting women, blue-collar voters and Hispanics. But in most contests the candidates failed to make such inroads (with Obama for instance falling short among these three groups respecively in NH, OH and CA) and thus the demographic logic was respected.

In Pennsylvania, demography favors Hillary Clinton; yet Obama has a clear shot at upsetting that logic. Whether or not he is in a position to win, he certainly could keep Clinton's winning margin dangerously low. But he will not do so by winning over Clinton's core constituencies but in winning the turnout war, with signs that blue-collar and Catholic voters might stay at home. If a large number do, Clinton will clearly be in danger. Not to mention the problem that even a 10% victory is not likely to do that much for her in the long-term.

Democrats should worry, however, for it will be difficult for them to recapture the groups of voters who might stay at home tomorrow; the fact that they could lose their enthusiasm in Clinton but not migrate to Obama suggest they are prime pick-up targets for the McCain campaign.

Clinton adds a superdelegate and strong polls... but is even 10% enough?

Now in the final days of the Pennsylvania campaign, Hillary Clinton got the support of a superdelegate today, Ohio Representative Tom Ryan. This is her second endorsement from Ohio in as many days, as Rep. Betty Sutton endorsed her yesterday afternoon. This is quite an unexpected double-get considering how slowly Clinton has been amassing superdelegates since February 5th, especially when we add to this the 2 superdelegates she obtained on Thursday when the New Jersey Democratic Party settled on three of its add-ons/PLEOs.

However, Barack Obama also received some good superdelegate news lately when Maryland Governor O'Malley announced (contrary to most people's expectations) that he would schedule a special election in the vacant MD-04 district. The incumbent Democrat (an Obama endorsee) had announced he would resign after losing to Donna Edwards in the Democratic primary; it appeared that the seat would be left vacant for the rest of the year and that Obama would lose a convention vote. But O'Malley decided a special election would be held after all on June 17th, with the winner automatically becoming a superdelegate. Donna Edwards, the overwhelming favorite to win on June 17th, has already endorsed Obama.

Meanwhile, 3 new Pennsylvania polls released this afternoon bring good news to Hillary Clinton and confirm that a double-digit victory is within her reach on Tuesday. Notice, however, that her trendline is mostly negative:

• The fifth installment of Zogby's tracking poll finds the New York Senator leading 48% to 42%. Zogby adds that, in Sunday's polling sample alone, Clinton leads 53% to 38% and that this is due to more than a one-day anomaly, as undecideds are breaking towards her.
• You might remember that we discussed this morning Clinton's biggest potential to improve her numbers as her own base looks to be wavering more than Obama's, suggesting that she has more room to grow in the final 48 hours.
  

• The final ARG poll, meanwhile, shows Clinton leading by a solid 54% to 41%. The previous poll released last week showed her leading by 20% -- in a poll that was widely described as an outlier; their previous survey (the week before) showed a tie at 45%, so ARG has shown some pretty wild swings.
  
• ARG's internals show Obama trailing 2:1 among white voters and by 23% voters among voters with no college degree; if those numbers hold on Tuesday, except a new round of criticism as to Obama's electability problems.
  

• Strategic Vision, finally, finds a 47% to 40% Clinton lead, here again a slight tightening as she was ahead 49% to 40% in the previous poll.
  
• In the general election, Strategic Vision once again finds McCain beating both Democrats, 46% to 42% against Clinton and 48% to 40% against Obama.

• [I had included the final SUSA poll here, but it looks like that was a very old poll I somehow linked too. I apologize for that, this is what happens after this many primaries in this little time. We will probably get SUSA's final poll tomorrow.]
  

With less than 36 hours before the polls open in Pennsylvania, these are among the final polls we will be seeing and they suggest that we are in for a wild race to the finish and for Election Day suspense. Clinton has managed to remain favored to win and she has consistently led throughout the 6-week campaign -- something she had not managed to do in Ohio and Texas. This is an impressive feat in and of itself considering how close her opponent is to capturing the nomination. That Hillary is capable of maintaining her coalition under such conditions is remarkable.

The problem she faces on Tuesday, of course, is that she needs not only a win but a comfortable one. And on that point the polls are in disagreement and they are simply too many undecideds to really know what will happen. One thing that is certain, however, is that the "good result" that Clinton is aiming for is a double-digit victory. While there is no question that that would be a strong result, it is important to keep in mind that the Pennsylvania primary started with the presumption that Clinton would need a much bigger victory to keep herself relevant. After all, given her precarious position in the race, this is no longer a question of beating subjective expectations but of (1) achieving the very objective result of moving pledge delegate totals and (2) changing the fundamentals of the race.

Clinton's victories on March 4th were strong enough to keep her in the race but they fulfilled neither of these latter two conditions. A 10% victory in Pennsylvania would do just about the same thing; it would allow Clinton to pass a threshold of survival but without putting Obama's dominance seriously in danger.

Catholic and blue-collar turnout: Two PA polls show Clinton narrowly leading, with more potential to expand

With Pennsylvania polls showing all kinds of trends and results over the past week, the similarities between today's Mason-Dixon/McClatchy and Zogby polls are striking:

• Zogby's tracking poll shows Clinton narrowly ahead, 46% to 43% -- a slight improvement for Obama who led 46% to 44% in the Saturday subsample alone (naturally, any subsample has a large margin of error).

• Meanwhile, the McClatchy poll shows Clinton leading 48% to 43%. Obama gets 83% of the black vote, but only 33% of the white vote.

Both institutes, however, point out that undecided voters or those who are unsure about whether they will go to the poll are more likely to break in Clinton's favor, implying that Hillary has a larger potential than Obama to move numbers in her favor. Zogby, first, points that the largest number of undecideds are whites and more precisely Catholics. Since Catholics break very heavily in Clinton's favor, Zogby concludes that, if the more undecided voters go to the polls on Tuesday the better Clinton will fare.

MSNBC's Chuck Todd offers a very similar analysis of the McClatchy poll. He notes that one of the highest proportion of undecideds is found in the rural areas of the state, especially in the "T" region of the state, and those are among Clinton's strongest areas. Overall, Todd implies that more blue-collar voters are suggesting that they are unsure of what they will do, and that obviously leaves more potential for Clinton than for Obama. Todd concludes that, "The clues inside the numbers indicate this is her race to lose and that her lead could expand."

Simply put, then, Zogby and Mason-Dixon find that there is a question mark about the turnout in rural areas, among blue-collar Democrats and among Catholics; the higher that turnout is, the better for Clinton. In a way, this is reassuring for Clinton, implying that it will be harder for Obama to climb into a lead on Tuesday since it appears that he has reached his full potential among his core constituencies.

Yet, it is also a clear sign that Clinton's electoral coalition is eroding, with some of the groups that have been solidly behind her since the first contest in January threatening to stay home and showing signs of tepidity. For much of the past 3 months, Obama and Clinton's base have been remarkably solid and equally enthusiastic -- which is what has created the impasse the Democratic Party finds itself in.

That Clinton's constituencies are now looking to be less enthusiastic and wavering is thus very significant. But it is not necessarily very surprising: Obama is very close to clinching the nomination, and the general election campaign between him and McCain has in many ways already started. Meanwhile, Clinton has to justify her continued presence in the race. It is thus not surprising that many of her supporters are not as committed to her campaign as they would have been if the race had been tighter; many are perhaps reluctant to support her if it could mean prolonging the campaign.

The fact that those groups of voters are taking refuge in the "undecided" column and in absenteeism rather than migrate to Obama, however, does suggest that the Illinois Senator will have to fight for their vote in the general election. Zogby notes that these groups of voters also contain the highest number of Democrats who say they will vote for McCain in the general election.

The Pennsylvania campaign: Revisiting expectations

3 days from the Pennsylvania primary, the Democratic electorate appears more volatile than it has been all year. With polls showing all sorts of lead and attributing momentum to both candidates, I am at a loss to say what the expectation is, first and foremost because it is too early to tell whether ABC's debate had an impact on voters. After all, a number of debates have moved numbers significantly at the last minute throughout the year, and polls have not been very good at picking that quickly.

Yesterday's Rasmussen poll, taken entirely after the debate, showed Clinton dropping 6%. Zogby's tracking poll released today, however, suggests that Clinton is moving upwards ever so slightly: She is leading 47% to 42%. And then there is Gallup's national tracking poll which shows Clinton suddenly taking the lead against Obama, 46% to 45%. That's a 12% gain in 4 days, and an 8% gain in 2 days. In other words, Clinton has picked up a significant amount of ground since the debate, with Gallup adding that, "Clinton has been significantly higher in both of these post-debate nights of interviewing than in recent weeks."

The stakes of Tuesday's voting are very high: Clinton wants to keep her candidacy alive, and Obama wants to demonstrate that he can be competitive among blue-collar voters and rural Democrats; even if he loses, success on that front would deprive Clinton of one of her best arguments. Complicating matters is the question of just what Clinton needs to accomplish on Tuesday to remain in the race; we have discussed this question many times already, and I suggested that Hillary needs to win by at least her Ohio margin and demonstrate Obama's weakness among core Democratic constituencies.

But how do recent events affect these thresholds? Bittergate, followed by Obama's weak performance at the ABC debate, have lowered the expectations the Illinois Senator has to meet; most people expect him to have plunged because of those two events. The story of Obama's San Fransisco remarks forced Hillary Clinton to use all her arsenal in the run-up to Pennsylvania, much before she expected to have to do her final push to superdelegates.

April 22nd has thus become a test of Obama's resistance and appeal when until recently it was a test of Clinton's resistance. Consider, after all, that Obama is the overwhelming front-runner whose campaign never misses an opportunity to explain why it is impossible for Clinton to clinch the nomination. Despite that, he could be about to lose a large state like Pennsylvania after a 6-week campaign and overspending -- and that is quite extraordinary.

Yet, the burden to meet a potential is on Clinton again rather than on Obama. The problem is no longer "Will Obama get the result that suits a candidate who is as inevitable as he is or will the contest expose that Democratic voters distrust him?" Instead, the main question has become: "Will Clinton get the result recent events allow her to hope for, or will her failure to do so reveal that Obama is immune to these controversies?"

If Obama maintains himself at a decent level but loses, Pennsylvania will become a celebration of his resilience rather than a discussion of why he is unable to clinch the nomination.

The stakes are high, therefore, and the number of negative ads is rapidly increasing on Pennsylvania airwaves. The latest by Obama (watch it here) brings us right back to the early days of 2008, since the Illinois Senator is blasting Clinton's health care plan because it "forces everyone to buy insurance even if you can't afford it." This is, of course, a very old line by Obama and one that got him in a lot of trouble with many on the Left and with the Edwardses. But the fact that Obama keeps coming back to it -- and believes that the spot is a strong closer for him -- suggests that his campaign has enough indication and research that this attack line is helping him among voters.

Clinton, meanwhile, is airing her own attack ads and some even suggested that 100% of her spots in certain markets are negative; Marc Ambinder contests that number but there is no question that the proportion of negativity has increased in the past two weeks from both campaigns. How that will affect the vote of Pennsylvanians in the general election, of course, will only be determined in the weeks and months ahead.

A bad polling day for Clinton

It is ironic that, on a day in which most state polls look bad for Hillary Clinton, the most talked about survey is Gallup's tracking poll. After 11 days by a statistically significant margin, Hillary Clinton has moved within the margin of error, 47%to 44%. That's a 4% improvement since yesterday and a 7% improvement in the past 3 days -- implying the movement started before the ABC debate.

If this movement is confirmed in the coming days and if the debate is found to move numbers towards Clinton, it would naturally be a significant development. But the most important polls released today suggest that Clinton is declining. A Newsweek survey shows Obama gaining a tremendous amount of ground in the past month, now coming in 19% ahead of his rival, 54% to 35%. If a few more polls confirm the existence of this massive a gap, it will be increasingly difficult for Clinton to hold off the superdelegate flight.

But it is the three polls from the upcoming primaries that are the most threatening to her survival past Tuesday and past May 6th:

• A Rasmussen poll of Pennsylvania, taken entirely yesterday, shows Clinton leading by a tight 47% to 44%. In the previous Rasmussen poll, released on Monday, Clinton was up by 9%.

• Zogby's tracking Pennsylvania tracking poll, meanwhile, shows Clinton leading by a narrow 47% to 43%. Yesterday, she was only ahead 45% to 44%, so today's numbers are a slight post-debate improvement. Zogby adds, however, that "22% of Catholic Democrats and 18% of white Democrats choose McCain against Obama."

• Meanwhile, the most surprising poll comes from Indiana: At the beginning of the week, SUSA found Clinton leading by 16%. Today, a new SUSA poll (conducted entirely before the debate) finds Obama ahead by 5%, 50% to 45%.
  
• The two SUSA polls were not conducted with the same turnout model or for the same outlet; so the two surveys are not directly comparable. However, the poll is clearly bad news for Hillary; the LA Time's poll a few days ago also showed Obama ahead by 5%.

There have been two movements in Pennsylvania polls. First, Obama picked up ground and threatened Clinton's dominance; then, Clinton managed to rebound and create some space. Polls today and yesterday suggest that the electorate is very volatile right now and most anything could happen on Tuesday.

Complicating matters for Clinton is that she will have to immediately work to survive May 6th if she gets past April 22nd. North Carolina has been a lost cause for a while, but it looked like Indiana held premise for Hillary -- the LA Times and latest SUSA poll indicate that Clinton will not have an easy time ensuring solid results in three weeks.

Finally, we also got interesting general election polls results today:

• Newsweek's national poll shows both Democrats ahead of McCain by the same narrow margin, 48% to 44% for Obama and 47% to 43% for Clinton.
  

• In Indiana, SUSA also conducted a general election poll and found a much tighter contest than what we have been accustomed to in the state: McCain leads Obama 51% to 44%. He is ahead of Clinton 53% to 42%. Bush got 60% and 57% in 2000 and 2004.

• In Washington, SUSA (again) found both Democrats leading McCain by differing margins: Obama leads 53% to 40%, while Clinton is ahead 48% to 45%. There is a huge difference among independents: Clinton loses them by 3%, Obama wins by 21%.
  

• In Colorado, meanwhile, Obama leads 46% to 43% against McCain; Clinton is trounced 50% to 46%.

• Finally, a Quinnipiac poll of New York shows McCain too close for comfort (just as yesterday's poll from SUSA) with Clinton ahead 49% to 37% despite only getting 65% of the black vote; Obama is held in single-digits, 47% to 39%.

One of the most consistent electability results is that Obama is much stronger than Clinton among independents out West -- Oregon, Washington and Colorado have consistently shown Clinton faring much worse than Obama. In fact, Colorado is one of the first states the Obama campaign they can take away from the Republican column; it is at the top of Obama's "alternative electoral map" that does not go through states like Ohio and Florida. On the other hand, Obama is more in danger in some of the bluest states, with SUSA yesterday showing him leading in single-digits in CA, NY and MA -- a dangerous finding confirmed by today's Quinnipiac poll.

Thursday polls: Are Democrats at risk in California?

Slow polling day today, as the bulk of the week's Pennsylvania polls were released earlier this week. That left the Thursday stage to Zogby, who released a poll from the Keystone state:

• Zogby shows a toss-up, with Clinton edging out Obama 45% to 44%. This makes Zogby only the second pollster with PPP to find this tight a race.
• In an interesting internal number, Obama is slightly ahead in union households. If he can hold on to that on April 22nd, it would be a great talking point for him to diffuse worries that he would be weak among blue collar voters.
  

We will naturally have to wait a few days for polls to start registering whatever impact yesterday's debate might have had on voters; in fact, Clinton's negative advertisement hitting Obama for his San Fransisco remarks only started airing at the beginning of this week, so the environment of the Pennsylvania campaign changed dramatically this week. That, too, can probably not yet be picked up by polls. It is of course unclear who all of this might help, but we will have to keep a close eye on poll trendlines in the next 5 days.

Meanwhile, two general election polls are of interest:

• In California, Rasmussen finds McCain competitive. He trails Obama 50% to 43% and Clinton 47% to 42%. Last month, McCain trailed Obama by 15% and Clinton by 7%.
• Obama at least benefits from a strong favorability rating (58%), whereas McCain at 48% and Clinton at 47% are weaker in this respect.

• In Colorado, a poll conducted by Republican polling firm TargetPoint Consulting shows McCain trouncing both Democrats, 51% to 39% against Obama and 52% to 40% against Clinton. I do not particularly trust this poll, and given its partisan nature take the results with caution (which also contained Senate numbers showing both candidates tied).

Colorado is an important state for Obama, one of the red states that is not traditionally at the forefront of Democratic preoccupations that the Illinois Senator thinks he can make competitive this year; the previous poll from the state (released by Rasmussen) shows him tied with McCain.

The worry for Democrats, of course, should come from California: They cannot afford to have this state even be competitive. Every day the Democratic nominee spends in California, every dollar that is spent on its expensive airwaves is that much less to go on the offensive in Ohio, Colorado and Virginia. A Democrat cannot enter the White House without California's 55 electoral votes, so McCain's staying in mid-single digits against both Democrats has to be a worrisome signs that the Arizona Senator has appeal on the coasts and appeal among Democratic-leaning independents who keep Democrats comfortably ahead in states like California.

Pennsylvania polls remain all over the place

No matter how often polls have been discredited throughout this cycle, it seems to me that we could usually rely on a certain consistency. In Ohio and Texas, Obama was clearly gaining in all polls before surveys released in the final 3 days showed the start of a Clinton comeback; on Super Tuesday, the trend was clearly towards Obama. Even in New Hampshire, where the final polls were so awfully off, I have always believed that the polls cannot be blamed but it is the media who didn't know how to read them given the 5 day window of NH campaigning and the last minute shift towards Clinton.

In Pennsylvania, however, polls show different trends, different momentum and different leaders. After SUSA and ARG, and even Strategic Vision and Rasmussen, showed Clinton solidly ahead yesterday, two new polls paint a picture:

• PPP finds Obama ahead, 45% to 42%. Last week, Clinton was ahead by the same margin. The poll was conducted Monday and Tuesday.
• Two weeks ago, PPP has already released a poll showing Obama ahead -- but no other institute has ever shown him leading.

• Franklin & Marshall University are more reassuring towards Clinton, who remains ahead 46% to 40% (49-42 with leaners).

So what is the effect of bittergate? There is little evidence of any effect at all, with different polls going in different directions. And it cannot be ruled out that the controversy is backfiring on Hillary Clinton. If some Pennsylvania voters had concluded that Obama had basically won the nomination but still wanted to cast a ballot for Clinton, her willingness to air an attack ad on this issue and to criticize Obama as harshly as she has could make some people think that it is too dangerous to prolong the primary. On the other hand, Clinton's ad just started running on Monday and the media is keeping the story alive, so there is only so long a politician can not take a hit over something that is repeated so often.

Thankfully, North Carolina polls are a bit easier to navigate. Two new polls today confirm the conventional wisdom that this state is favoring Obama:

• An ARG poll shows him leading 52% to 41%, which is a two point improvement for Hillary since last week. Clinton narrowly trails among women (50-44), crushes Barack among whites (57-33).
  

• Meanwhile, Insider Advantage finds Obama ahead 51% to 36%. The pollster adds that there has been little movement since the end of March, suggesting bittergate has not caught on the imagination of North Carolinians.
  
• He adds, as a warning to Clinton, "The makeup of the white Democratic vote in North Carolina is heavily populated with voters from ‘Research Triangle’ universities-based region." That is not the type of electorate Clinton ever hopes for...

These polls give us an idea of how difficult the road ahead is for Clinton. She needs to post strong numbers in PA and she then immediately needs to get equally solid results in North Carolina and Indiana. A double-digit lead in the second largest state left to vote (after PA) will simply not cut it. The problem for Hillary is that the NC electorate is in some ways similar to Virginia's, with a significant proportion of the white-collar whites (though not as much as VA, allowing her a route to stay competitive).

Finally, the last poll of this morning comes to us courtesy of the Washington Post, which finds that Obama is now ahead of Clinton 51% to 41% nationally; against McCain, Clinton trails 48% to 45% while Obama is ahead 49% to 44%. Could it be that national numbers are more important than ever since, for the first time, they have an audience (superdelegates)?

As more ads come in, polls from upcoming primaries are all over the place

There are a lot of polls coming in today from many different primaries, and we are forced to conclude that the numbers are all over the place, particularly in Pennsylvania and Indiana where it looks like anything could happen.

Pennsylvania, first, where 3 polls this morning showed Clinton up 14, 5 and 9. Two new surveys released this afternoon show Hillary leading by single digits:

• LA Times/Bloomberg poll shows Hillary leading in Pennsylvania 46% to 41%. The survey was conducted Thursday through Monday, so partly before bittergate.

• Strategic Vision, meanwhile, finds Clinton increasing her lead by four percent in a week; she is now ahead 49% to 40%.

My analysis of the Keystone state's primary does not differ form this morning's, when I explained that it is difficult to assess what effect Obama's comments have had, but that the Clinton campaign at least has to be happy that it has stopped the bleeding and the steady plunge it went through for 2 weeks or so. This is not because Obama isn't trying, however: he is outspending Clinton 5:1 on Pennsylvania cable TV (!), which is truly a huge proportion.

Today, Obama launched his response ad (you can view it here) to the attack ad Clinton started airing yesterday centering on Obama's remarks. The response ad shows Clinton being jeered when trying to bring up an attack line against her rival, and then goes on to say that we should be looking to work on concrete problems and leave politics as usual behind. The ad is a rather typical response to an attack ad by a front-runner. Indeed, front-runners have nothing to gain from attacking back and they try to portray themselves above the fray (which is exactly what Clinton did for months in the fall when Edwards and Obama were going after her). It is also interesting that the ad never mentions what Clinton is attacking Obama on, so Obama is not using this opportunity to defend himself against criticism.

If Clinton survives Pennsylvania (and with a couple of polls today showing her on the verge of the "Ohio threshold," she is more than ever in a position to), the race will go on to North Carolina and Indiana (on May 6th). A few weeks later would come Kentucky (May 20th). We got a poll from each of these 3 races today, with good news for both candidates:

• In North Carolina, which Obama is heavily favored to win, the LA Times poll shows him leading 47% to 34%.

• In Indiana, where there is less of a defined favorite, Obama posts a 50% to 45% lead (still LA Times).

• In Kentucky, finally, SUSA finds that Clinton demolishes Obama 62% to 26%... much more than a 2:1 margin. SUSA's previous poll already had Clinton way up, 58% to 29%.

The most interesting numbers among this set is the Indiana survey. However much the primary has no clear favorite, the two only polls we have gotten from the state have shown Clinton lead. They have both been released from SUSA, and the latest, yesterday, had her leading by 16%. A conventional wisdom was starting to set in, then, that Indiana was Clinton's to lose. The LA Times reestablishes some balance in our expectations. But it is clear that Clinton needs a comfortable win here on May 6th; she will need it to offset the likely loss of North Carolina.

Finally, an interesting question in Pennsylvania is what effect the prolonged primary has on the general election; there have been a number of polls suggesting that Democrats are improving their position in the Keystone state against McCain and that their millions spent on tv are helping them in the general. Strategic Vision's poll, however, suggests the exact opposite, as this is the second week in a row that McCain is gaining. Last week, Clinton was leading by 6% and Obama trailing by 5%. In this week's poll, Obama is led 49% to 39% and Clinton trails 47% to 44%.

Pennsylvania polls trickling in

With 3 new polls of the Pennsylvania primary out today, it is very difficult to conclude what effect -- if any -- Obama's "bitter" remarks have had on the race. This is especially so because the Quinnipiac poll, which will be the most cited today because it has one of the best reputations out there, was conducted from April 9th to April 13th, so a very large portion of its interviews were conducted before the controversy erupted:

• In the Quinnipiac poll, Clinton leads 50% to 44%. This is exactly the same margin as last week.
  
• Quinnipiac provides the numbers for the group it calls "Reagan Democrats," among which Clinton leads 55% to 40%. This is obviously the group that Clinton wants to seduce to demonstrate Obama's general election weakness.
• The big worry for Obama: Only 50% of Clinton supporters say they will vote for Obama in the general, versus 26% who say they will vote for McCain. Among union households, it falls to 47-32. Clinton fares better among Obama voters, 70% of whom would vote for her, including 74% of Obama supporting union households.
  

• The Rasmussen survey, meanwhile, finds Clinton picking up some ground. She now leads 50% to 41% -- up from a five percent lead last week.
  
• The poll was conducted entirely Monday. 75% of Pennsylvanians say they heard about the remarks, and only 35% said they agreed -- but 59% of Obama supporters.

• Finally, Survey USA released its weekly poll, finding Hillary Clinton up 54% to 40%. This is actually a slight decline from last week's poll for Clinton; in that survey that became widely discussed, criticized as an outlier or as the start of a trend, Clinton led by 18%.

The abundance of polls released since late next week confirms, at least, that the race is no longer tightening like it did for much of the end of March and early April. At one point, it only seemed like a matter of time before Barack would overtake Hillary but the New York Senator rebounded in most polls since -- starting before bittergate.

This makes it especially difficult to assess whether any trend discovered in these polls has anything to do with the coverage of Obama's remarks. Even polls showing a Clinton improvement are inscribing themselves in the continuity of last week's polls, rather than showing a drastically new trend.

Clinton is doing her best to ensure that there is a new trend and as dramatic as possible an acceleration of the movement towards her. She is now running an attack ad against Obama for his "cling to" remarks. And today, she is assembling a team of 100 Pennsylvania Mayors for a collective endorsement; an event is being organized in Harrisburg at which many of them will appear, and some "small town" mayors will hit Obama over his comments about small town voters. It does look, however, like many of these mayors already support Clinton. This is clearly not meant to announce new endorsements as much as adding hype to bitter-gate, making sure the media covers such as a large event and reaching more voters.

Post-"bitter" Pennsylvania expectations: First poll has Clinton gaining

The first poll of Pennsylvania since the story over Obama's "bitter" comments broke was released today by ARG. It was conducted April 11th to 13th, so a part of the interviews were conducted before any controversy erupted (it is unclear what proportion):

• Hillary Clinton has jumped up to a 57% to 37% lead, almost entirely due to the female vote, which she leads 64% to 31%. The other surprising numbers is how much Clinton leads by among whites, 64% to 29% -- much than we are accustomed to seeing.
  
• A week ago, Obama had pulled within a tie, at 45% -- meaning this is a 20% jump in Clinon's favor.

It is too early to suggest that Obama's plunge in ARG's new Pennsylvania poll is due to the "bitter" controversy. Remember, Clinton started to rebound in a number of polls last week (PPP, SUSA and Insider Advantage), so ARG's survey inscribes itself in their community rather than break with all preceding trends to point to a dramatic turnaround. Consider also that the poll was taken immediately following heavy coverage of the issue, the time at which there will be the most impact in polls (if there is an impact to be had at all). We will have to carefully watch the polls that will be released in the coming days.

However, and as I just mentioned, there is now plenty of polling data pointing to a Clinton rebound. The size of Clinton's surge in the ARG poll might be an outlier, but Pennsylvania numbers have been moving away from Barack in the past week. Enough for Hillary to survive April 22nd? One consequence of the "bitter" controversy is that it ups the stakes of the Pennsylvania showdown. Clinton surrogates are now openly arguing that Obama is too weak among blue-collar voters to beat McCain in the fall, and they are hoping that Obama's comments resonate with superdelegates.

This means that the Pennsylvania vote will be watched very closely for evidence as to how weak Obama truly is among blue-collar voters and how much his San Fransisco comments hurt him. If the final numbers resemble this ARG poll, it could lead to very difficult days in the Obama campaign. But if he comes close to Clinton, he will claim to have put the controversy behind him; Hillary's electability argument will then take a very serious hit. Obama is clearly aware that he needs to bolster his numbers among blue-collar voters. His latest ad (which you can view here) features Senator Bob Casey, most popular among downscale whites.

In many ways, then, "bitter-gate" as some are now calling it could prove an opening for Obama. Consider that, as of last week, Clinton's lead had evaporated and Obama was playing catch-up remarkably well; but, with that improvement, expectations had changed as well: Just like in Ohio, the question had become who would win Pennsylvania -- rather than by how much. This change benefited the Clinton campaign which is seeking to stay alive no matter how small a victory it obtains on April 22nd.

But after 3 days of media hysteria over bitter-gate, the expectations are not only that Clinton will win the Keystone state but that Obama is badly hurting. A single-digits victory by Hillary, while it might have looked like a big relief as of last week, now would appear like a missed opportunity -- which is obviously welcome news for Barack.

One last note for now on bittergate, via Ben Smith: Speaking in Pennsylvania, Clinton said that "I don't think he really gets it that people are looking for a president who stands up for you, not looks down on you. And after seven years of Americans feeling invisible to this President, to President Bush, it's time that we level the playing field and begin acting like Americans again and that we roll up our sleeves and we get to work." (video here) Having criticized McCain's first ad with the tagline "this American president" for seeking to portray Obama as un-American, I have to at least point to Clinton's quote, for she does seem to be saying that Obama's elitist behavior is un-American, just like John Kerry was accused of being "French" in 2004 for not understanding blue-scale voters. But since the formulation is kind of confused, what do you guys think "acting like Americans" is referring to here?

Are Dems getting a boost in Pennsylvania's general election?

Confusion continues to surround the state of the Pennsylvania primary, with different polls attributing momentum to both candidates in the past few days. After appearing to rebound in a number of polls two days ago, today's two polls suggests the race is heading towards toss-up status:

• Zogby finds Clinton to be leading Obama by 4% in a race characterized by a large gender gap and a larger racial gap.
  

• Temple University shows a sensibly similar result, showing Clinton up 47% to 41%. Clinton gets 55% of the female vote and 32% of the male vote.
• This poll provides interesting data points to measure the claim that Obama brings more disaffected voters to the polls: While 27% of Obama supporters say they have never voted in a primary before, versus 18% of Clinton supporters, only 55% of Clinton's say they follow public affairs most of the time, versus 63% of Obama's backers.
  

The wide disparity that polls are registering is making it impossible for the campaigns to set expectations. Besides the obviously significant results (an Obama victory, a Clinton landslide), how will intermediary ones be perceived? A single-digits victory by Clinton will undoubtedly be perceived as insufficient for her to dent Obama's growing inevitability, while a low double-digit victory would result in some heavy spin by both campaigns. Considering that the conventional wisdom has now incorporated the tightening of the race, simply crossing the "Ohio threshold" could prove enough for Hillary to claim a resounding victory. What do you all think -- what margin does Clinton need to survive beyond April 22nd?

Meanwhile, Rasmussen released another poll from Pennsylvania today -- this one testing general election match-ups. And it shows some solid results for Democrats:

• Clinton beats John McCain 47% to 38%, practically the same margin claimed by Obama (47-39).
  
• A month ago, McCain narrowly edged out both Democrats. McCain's favorability rating has declined to 50%, while both Democrats have marginally improved theirs (57% for Obama and 53% for Clinton).

Pennsylvania numbers have been particularly disastrous for Democrats for much of the year; McCain often comes in stronger in the Keystone State than he does in Ohio. At the beginning of the Pennsylvania campaign, however, I explained that the primary dragging on was not necessarily a bad thing for the party and that "a 6 week long campaign in Pennsylvania could go a long way towards locking that state in the Democratic column come November."

News that Obama had been spending millions every week to cover the state's airwaves with ads only reinforced this argument: Obama has been blanketing one of the most crucial general election swing states with positive spots introducing himself, giving himself a head start in the general election race. Naturally, the increasing polarization of the Democratic primary makes it very possible that the eventual nominee will have trouble bridging the rifts among Pennsylvania Democrats, but I maintain that the PA campaign has been too civil for Democratic cohesion to be truly endangered.

Rasmussen's poll shows a big shift in the past month -- though we obviously need much more than one poll to conclude that there is evidence of a trend. The two other institutes that poll the state with any regularity are Strategic Vision, which recently showed slight movement in McCain's favor and Quinnipiac, which showed both Democrats slightly improving their numbers from mid-February to early April. A few more general election polls will hopefully be released soon after the April 22nd primary to give us a better idea of how the Democrats' position in Pennsylvania was impacted by this 6-week campaign.

In other polling news, Rasmussen finds that, as expected, Democrats should not count Louisiana's 9 electoral votes and a national poll suggests the general election is increasingly tight:

• In Louisiana, McCain crushes Clinton 58% to 36%. Obama is a bit more competitive, trailing 52% to 41%. Hillary's favorability rating is a dismal 37%, far from McCain's 64% and Obama's 53%.

• The AP-Ipsos national poll suggests that Obama's lead against McCain has collapsed since February 24th, when he led 51% to 41%. Now, the two are tied at 45% -- with McCain's taking a crucial lead among independent voters. Hillary Clinton, who led McCain by 5% in February, remains ahead 48% to 45%.
  

Thursday polls: First numbers from Puerto Rico!

Two important primaries polls were released this morning -- including the very first ever from the potentially decisive contest in Puerto Rico. First, however, comes Pennsylvania and its April 22nd primary. It's become increasingly difficult to figure out where things stand in the Keystone state, with SUSA and Insider Advantage (and, to a lesser degree, PPP) showing Clinton on the rebound while other pollsters still find her lead declining:

• Today, TIME magazine released its own poll, finding Clinton ahead 44% to 38%, 49% to 41% once leaners are included.
• The poll's internals are expected: Clinton has a big lead among white voters (51%-30%) and particularly among white women; Obama is ahead among blacks 80% to 8% -- and he still has room to grow here, considering he was routinely reaching 90% of the African-American vote in late February and early March. Also, 26% of Clinton voters say they would be "more likely" to vote for McCain is Obama is the nominee.

Both campaigns are now battling in a ferocious game of expectations; the Clinton campaign claims it is a miracle she is even ahead given how massively she is being outspent, an argument they already used in the run-up to Ohio and Pennsylvania. Meanwhile, Obama's camp laughs that off and explains that a single-digit loss would be a great victory. Unfortunately for Hillary, Obama is right to say that this is too late in the game for "just a victory" to help Clinton much. She needs to both get a meaningful pledged delegate lead out of the state and trounce Obama by enough of a margin to instill doubts about his candidacy. The good news for the Clinton campaign is that polls suggest she has room to grow back to the kind of lead she needs (just like in Ohio) -- and the seemingly constant stream of bad Pennsylvania polls has been interrupted.

Meanwhile, the first poll from Puerto Rico was released today! The primary is being held on June 1st and will award a large number of delegates that could be decisive if Clinton is hoping to cut Obama's lead. More importantly, Puerto Rico will almost close voting season (there will be two small votes on June 3rd) and the result could very well determine which campaign heads into the "off-season" with the most momentum:

• This new poll shows Hillary ahead 50% to 37% against Obama. The conventional wisdom is that this is Clinton's contest to lose and this margin might even look a bit underwhelming.

Puerto Rico could become an organization battle -- if the race goes that late. Bill Clinton already visited Puerto Rico this past month, and the state's establishment is mostly backing Clinton. In fact, Obama's main support in the state was Governor Acevedo Villa who was indicted last month of 19 counts of campaign finance violations, limiting the influence he could have had on the vote.

And you can bet the exhausted campaign and the media would love to spend a few weeks in Puerto covering the primary.

Update, in response to a comment by Jason: The "seemingly constant stream of bad PA polls" being interrupted has nothing to do with weekend polling versus weekday polling. Practically every poll over more than a week showed Obama rising and a number of institutes no longer see such a trend this week: SUSA, Insider Advantage and PPP all had Obama catching up last week and show the reverse movement this week. Also, most polling institutes go on the field week-to-week (say Monday to Wednesday; or Friday to Sunday) and they use the same turnout model. This is why so much is made of trends between the two last polls of one institute. The study you refer (I had mentioned it here) was only applicable to the Gallup and Rasmussen tracking polls which are in the field every single day, leading to comparison problems if one candidate polls better on a weekend.

Spending and polls: Where things stand in Pennsylvania

Now less than 2 weeks from the Pennsylvania primary, the key question is whether Barack Obama should take seriously his chances in the Keystone state. With every poll showing a tightening race, the Obama campaign was picking up confidence until SUSA's survey yesterday showed Clinton jumping back up to an 18% race. Since then, the question has become whether SUSA's poll is an outlier, and while most other polls show a much tighter race SUSA's track record and turnout model have been good throughout these primaries.

Today, three Pennsylvania polls were released and tell differing stories -- though 2 out of 3 show Clinton on the rebound:

• First came PPP which had published last week the first and only poll to show Obama ahead. Today, Clinton has taken back a narrow lead, 46% to 43%. She leads 53-36 among whites, Obama 74-17 among blacks.

• Second, a Strategic Vision survey shows Clinton ahead 47% to 42%. Last week, her lead was slightly bigger, 49% to 41%.

• Update: A third poll from Insider Advantage confirms yesterday's SUSA poll as Hillary has picked up to lead 48% to 38% -- roughly the threshold between a respectable showing and one that Obama can use to push her out of the race. Last week, Clinton's lead was reduced to 2%.
  

So where do things stand in Pennsylvania, and how much upward potential does Obama have? This question goes much beyond poll-centered musings: It is a crucial strategic question for the Obama campaign, which has to decide how much they should invest in the state in the coming 12 days and how long should their candidate spend campaigning in the state. If they decide that he has little chance of overtaking Clinton, they could resort to the strategy they had adopted for the first two weeks of the PA campaign: Ignore the primary and make it as irrelevant as possible, depriving Clinton of some of the momentum she would get out of a big victory.

But the Obama campaign is no longer holding back and now seems to be going all-out, at least on airwaves. He has outspent Clinton 3:1 for now, and he is currently spending a stunning $2.2 million a week on Pennsylvania airwaves, twice the total of his rival. And it is certainly not because Clinton isn't doing her best to stay in touch; in fact, she just launched 5 TV ads a few days ago, including one in Spanish.

The SEIU is doing its best to close the gap in Pennsylvania by canvassing on behalf of Obama and has decided (via Ben Smith) to spend $740,000 in its effort -- an even bigger financial hammer on Clinton's efforts. This should also help Obama improve his numbers among the constituency he needs the most on April 22nd -- blue collar voters. Obama's biggest preoccupation, beyond even winning, is demonstrating that he has appeal there as one of Clinton's major arguments to superdelegates is Obama's weakness among that group. This is evident by the type of ads Obama is running in Pennsylvania; presenting his latest clip, TPM notes that ads featuring big rallies have been replaced with "him talking sense to a small roomful of nodding voters."

In non-Pennsylvania primary news, note a new poll from North Carolina released last night by SUSA which shows Obama leading 49% to 39%, with a 22% Clinton lead among whites. This is the lower end of Obama's winning margins that we have seen in the past weeks; it is safe to say that any double-digit Obama victory here would be dangerous for Clinton, especially if it is accompanied by inroads among white or downscale voters by Obama.