Pennsylvania primary predictions

editor's note
Crystal ball gazing for elections is always hazardous, even when the complicating factors present in today's Pennsylvania's Democratic primary are absent. But I asked AT's political mavens to predict what they think the outcome will be when the votes are tallied tonight.

Richard Baehr, chief political correspondent:

I think the late movement is toward Clinton, as it was in Ohio, where she won by 10%. Her lead in the RCP composite is only up one point from 5% to 6%, but I think her victory may be bigger than that-- more in the 7 to 10% range. If she beats the high end of that range, it might change the dynamic of the race.  Anything less than a 6% victory will lead to even more calls for her dropping out.

But if she wins Pennsylvania, she will have won every large state (18 Electoral College votes or more) except Obama's home state of Illinois.   With polls showing McCain competitive with Obama in reliably Democratic California and New York, and other polls showing Clinton running better against McCain in the states that are always close (Pennsylvania, Michigan, Florida, Ohio), another big  win for Clinton should be a factor with super delegates.

But this reasoning does not mean it will be, due to the misleading media reports portraying Obama as far ahead in the popular vote (not true -- just over 2% lead, 1% with Florida)  and in the elected delegate race (6% lead for Obama, bigger than the popular vote margin due to his big wins in states with low turnout caucuses).  The major media are completely in the bag for Obama, in a pretty shameless fashion at this point reflected in their circling the wagons to protect their candidate after the ABC debate. 

Rick Moran, associate editor:

All depends on turnout. A big turnout of white voters in the western and northern parts of the state along with suburban vote could give Clinton the double digit win she's looking for.
But I think she'll fall just short of that. Prediction: Clinton 54 Obama 46.

Thomas Lifson, editor:

I do not trust the polling data in this race. Ever since race entered the equation, thanks to Bill Clinton's statements in South Carolina comparing Obama to Jesse Jackson, and especially since Rev. Wright became an issue, I suspect that many people are not telling strangers who telephone them how they plan to vote. That is the so-called "Bradley effect." 

Drudge says  that Hillary's internal polling shows an 11 point victory coming. This is plausible, given the probable impact of the "cling" gaffe on San Francisco's Billionaire' Row and Obama's poor debate performance. But I suspect that the internal polls may be optimistic, since the Hillary camp may be getting what it wants to hear from the pollsters whom it pays. Michelle Cottle has just written  about the incredible chaos in that campaign operation. At this point, the level of dysfunction could extend to polling.

If I have to pick a margin, I would say 10%. But if some Philadelphia precincts record 100% turnout as they have in the past, perhaps Obama will lose by as little as 7%.

Of course, the really significant electorate is the superdelegates. They have to be thinking about the reaction of the African-American constutency if they dump Obama. My guess is that they will not choose Hillary in sufficient numbers to give her the nomination, even if she wins by double digits.
Crystal ball gazing for elections is always hazardous, even when the complicating factors present in today's Pennsylvania's Democratic primary are absent. But I asked AT's political mavens to predict what they think the outcome will be when the votes are tallied tonight.

Richard Baehr, chief political correspondent:

I think the late movement is toward Clinton, as it was in Ohio, where she won by 10%. Her lead in the RCP composite is only up one point from 5% to 6%, but I think her victory may be bigger than that-- more in the 7 to 10% range. If she beats the high end of that range, it might change the dynamic of the race.  Anything less than a 6% victory will lead to even more calls for her dropping out.

But if she wins Pennsylvania, she will have won every large state (18 Electoral College votes or more) except Obama's home state of Illinois.   With polls showing McCain competitive with Obama in reliably Democratic California and New York, and other polls showing Clinton running better against McCain in the states that are always close (Pennsylvania, Michigan, Florida, Ohio), another big  win for Clinton should be a factor with super delegates.

But this reasoning does not mean it will be, due to the misleading media reports portraying Obama as far ahead in the popular vote (not true -- just over 2% lead, 1% with Florida)  and in the elected delegate race (6% lead for Obama, bigger than the popular vote margin due to his big wins in states with low turnout caucuses).  The major media are completely in the bag for Obama, in a pretty shameless fashion at this point reflected in their circling the wagons to protect their candidate after the ABC debate. 

Rick Moran, associate editor:

All depends on turnout. A big turnout of white voters in the western and northern parts of the state along with suburban vote could give Clinton the double digit win she's looking for.
But I think she'll fall just short of that. Prediction: Clinton 54 Obama 46.

Thomas Lifson, editor:

I do not trust the polling data in this race. Ever since race entered the equation, thanks to Bill Clinton's statements in South Carolina comparing Obama to Jesse Jackson, and especially since Rev. Wright became an issue, I suspect that many people are not telling strangers who telephone them how they plan to vote. That is the so-called "Bradley effect." 

Drudge says  that Hillary's internal polling shows an 11 point victory coming. This is plausible, given the probable impact of the "cling" gaffe on San Francisco's Billionaire' Row and Obama's poor debate performance. But I suspect that the internal polls may be optimistic, since the Hillary camp may be getting what it wants to hear from the pollsters whom it pays. Michelle Cottle has just written  about the incredible chaos in that campaign operation. At this point, the level of dysfunction could extend to polling.

If I have to pick a margin, I would say 10%. But if some Philadelphia precincts record 100% turnout as they have in the past, perhaps Obama will lose by as little as 7%.

Of course, the really significant electorate is the superdelegates. They have to be thinking about the reaction of the African-American constutency if they dump Obama. My guess is that they will not choose Hillary in sufficient numbers to give her the nomination, even if she wins by double digits.