Is Pennsylvania Really In Play?

Once again, the Keystone State is teasing the GOP.


Republicans haven’t won Pennsylvania in a Presidential election year since George H.W. Bush managed to do it in 1988. Ever since then, there have been years where the Keystone State has teased Republicans into thinking they have a shot there, only to find that hope dashed come Election Day. This year seems to be one of those year and, right on time, the polls are starting to tease Republicans again.

First up, there’s a new Quinnipiac poll showing only a four point race between the President and Romney:

Gov. Mitt Romney has narrowed a 12-point gap with President Barack Obama and now trails the president 50 – 46 percent among Pennsylvania likely voters, according to a Quinnipiac University poll released today.

This compares to a 54 – 42 percent Obama lead in a September 26 survey by the independent Quinnipiac (KWIN-uh-pe-ack) University.

In today’s survey, men back Romney 54 – 43 percent, compared to a 49 – 48 percent split September 26. Women back Obama 57 – 39 percent, little changed from last month. White voters back Romney 53 – 43 percent while black voters back Obama 97 – 1 percent. White Catholic voters go Republican 56 – 43 percent. Voters with college degrees back the president 54 – 43 percent while voters without degrees are divided with 49 percent for Obama and 47 percent for Romney.

Only 7 percent of Pennsylvania likely voters say they might change their mind in the next 21 days.

“Gov. Mitt Romney is coming on strong in the Keystone State, especially among white Catholics,” said Tim Malloy, assistant director of the Quinnipiac University Polling Institute.

These results are mirrored in another poll by Muhlenberg College and the Allentown Morning Call:

The survey also squares with another recent Pennsylvania survey that reported Romney gains. A Muhlenberg College/Morning Call poll of likely voters, conducted from Oct. 10 through Oct. 14 with a margin of error of 5 percentage points, found Obama leading Romney by only 4 points, 49 percent to 45 percent. That was down from an 7-point advantage for the president from the same survey taken in late September.

Judging by these two polls alone, it would seem like the there’s some real movement toward Romney in the state. However, there are other recent polls that show a far different view of the state of the race. A recent Public Policy Polling survey has the President up by seven points, a Rasmussen poll has Obama up by five points, and a Philadelphia Inquirer poll has the President up by eight points. In fact, the only recent poll that seems consistent with the view of a tight race in Pennsylvania is a Susquehanna poll that shows only a two point gap between the candidates, a number that seems far too small. This leaves us with a RealClearPolitics average of five points in favor of the President, a lead but a smaller one than he’s had in the past.

I’m still skeptical of the idea that we’re seeing anything real from Pennsylvania this time around. As I’ve said before, this state has teased the GOP before with the hope that it could be captured. In the end, though, the overwhelming Democratic turnout from Philadelphia and its suburbs, and to a smaller extent Pittsburgh, overwhelms Republican voters from other parts of the state, and turns the state blue once again. There’s not really any reason to believe that this isn’t going to happen again this year. Moreover, there’s absolutely no sign that the Romney campaign believes that the state in play. They aren’t spending large amounts of money there, and we haven’t seen the same kind of attention paid to the state as we have to states like Ohio. That makes sense, of course, because even the large amounts of money that Romney has raised this year can only be spent in so many places, and it seems smarter to spend time in the Virginia, Ohio, Colorado, Nevada, and possibly Wisconsin than a state that hasn’t gone Republican in twenty-two years.  Barring some extremely unlikely developments, I expect Pennsylvania to go blue again this year.


FILED UNDER: 2012 Election, Public Opinion Polls, US Politics, , , , , , , , , , ,
Doug Mataconis
About Doug Mataconis
Doug Mataconis held a B.A. in Political Science from Rutgers University and J.D. from George Mason University School of Law. He joined the staff of OTB in May 2010 and contributed a staggering 16,483 posts before his retirement in January 2020. He passed far too young in July 2021.


  1. Fiona says:

    Agreed. Especially now that the state’s voter ID law, which the Republican majority leader admitted was designed to hand the state over to Romney, has been put on hold.

  2. PJ says:

    We all know, thanks to Smooth Jazz, that almost all pollsters are liberals wanting to reelect the President so they fudge their numbers.

    Quinnipiac, one of these liberal pollsters, now say that Romney is closing the gap in PA.

    Isn’t it obvious why? Obama has told them to do this so that Romney will burn money in a state that the Republicans can’t win.

    All liberal polls lie all the time, they just change the lies.

  3. EddieInCA says:

    There is virtually no spending by either party in Pennsylvania. Hence, it’s not close.

    I’m in Florida, and the radio stations I listen to are running about 3-1 ads Pro-obama vs. pro-Romney.

    Maybe Romney doesn’t want or need the sports talk audience?

    Television ads seem to be equally split here in South Florida.

  4. mattb says:

    Doug wrote:

    Moreover, there’s absolutely no sign that the Romney campaign believes that the state in play. They aren’t spending large amounts of money there, and we haven’t seen the same kind of attention paid to the state as we have to states like Ohio.

    Here’s the key fact. Until Romney reinvests in Pennsylvannia, its only in play in the minds of media analysts.

    As far as the narrowing of polls, it’s most likely the case. But at this point in the race, a five point difference (using the Rasmussen number) is very significant as we are no longer discussing “margin of error” given the general stability of the polls and the continued reduction in the number of undecided voters.

    The only hope Republicans have for Pennsylvania are two similar debate performances (from both candidates) will motivate a massive GOP gotv effort. Still given the state’s long history of going democratic in presidential elections, that has to be one heck of a gotv ground swell.

  5. Geek, Esq. says:

    It’s more of a symptom of Obama’s decline nationally.

    Can he really be up +4 in PA but +5 in OH? Nope.

    OH runs about 5.5 – 7.5% more Republican than does Pennsylvania. So, in general Obama should feel comfortable if he’s winning PA by about 8 and should be sweating bullets if his lead there is 5 or less.

    And, whaddya know, the polls in PA run from 4 -8 points.

  6. Geek, Esq. says:


    Either it’s not really in play, or it’s in play in the sense that it could be a margin padding state in a Romney wave, much like Obama padded his margin last year in FL, VA, IN, and NC.

    In 2008, McCain didn’t focus hard on protecting Indiana–correctly figuring that if Indiana were close enough to matter, he was screwed anyways.

  7. KariQ says:

    I think Pennsylvania is about as much in play as Arizona. At least there is a poll that shows Obama ahead in Arizona, even though everyone regards it pretty much as a joke.

  8. Just Me says:

    I am hard pressed to believe Obama is going to lose PA. Romney may finish closer in the state than previous GOP candidates, but I really can’t imagine Romney will win the state.

    I also think the fact that neither party is spending money in the state probably indicates that Obama feels safe in PA and Romney doesn’t think he can win it.

  9. Geek, Esq. says:

    Nationally, Romney is up around 2-4 points (Gallup likely voters and Daily Kos polls both show 50-46). PA is about +4 Democratic compared to the national vote, so Obama probably is barely ahead here.

    The question for both campaigns remains whether (a) they expect those numbers to hold and (b) whether resources spent in Pennsylvania would have a higher return than those allocated to Ohio and Wisconsin.

    Romney’s got the wind at his back, and in politics winners play offense and losers play defense, so there’s definitely an argument to be made for making a modest play.

    On the other hand, Bush got too greedy in 2000 and nearly lost the election because he went for a victory lap in California.

  10. C. Clavin says:

    Also of note today – SCOTUS refused to hear the case of Republican efforts at Voter Suppression in Ohio.

  11. KariQ says:

    @Geek, Esq.:

    Nationally, Romney is up around 2-4 points (Gallup likely voters and Daily Kos polls both show 50-46). PA is about +4 Democratic compared to the national vote, so Obama probably is barely ahead here.

    Disagree on both counts. Pulling out the two most Romney leaning polls and declaring them correct is a bad idea. Gallup’s LV model – and they’ve admitted this themselves – doesn’t work well until the week before the election so relying on them is especially risky. The national polls range from Romney +4 to Obama +3. The RCP average is Romney +0.6. If you want to round that off and say Romney +1, fine. But throwing out all the other polls is a good way to be stunned at the actual results.

    Pennsylvania polls range from Obama +2 to Obama +8, so it’s safe to say that Obama has a significant lead there.

  12. Tsar Nicholas says:

    Money talks and bullshit walks.

    If the campaigns are not spending money in PA then it’s not really in play. Although obviously Obama’s margin there will be far less than it was in ’08.

    Speaking of which, Geek, Esq. has touched upon the giant flaming neon elephant in the room, which amazingly enough apparently is not registering with the chattering classes. Obama in ’08 won PA by an overwhelming landslide margin and in so doing far outstripped his correponding margin in the national popular vote. PA was and still is far more of a Democrat state than, gulp, Florida, Virginia and Ohio and a few of the other key swing states. Now the RCP polling average is Obama + 5 in PA. And that includes a PPP survey, which shamelessly oversampled Democrats. Connect the dots……..

  13. bk says:

    Shorter a lot of people: Polls don’t matter, unless they are favorable to our side, in which case they matter.

  14. bk says:

    And I can actually say, without being sarcastic, that in this case yes – both sides do it.

  15. Geek, Esq. says:


    As Josh Marshall noted a few weeks ago, the side that’s taking issue with poll methodologies is likely to have a depressing November 6.

    The blind spot that team Obama had was that they never saw a need for a coherent big picture narrative. This is due in part to Obama’s non-ideological pragmatism and partly because they tend to lose the forest for the trees–everything is microtargeting, whether it be messages, demographics, or even which states to contest. They’ve never bought into the idea that this would be race that the entire nation would vote on–rather it was subgroups in battleground states they focused on.

    That was the real takeaway from the first debate–here was this supposed change agent whose only idea for the future was 100,000 new science teachers. He had nothing to tell the entire nation as to why his second term would be better than the first, or why his second term would be better than a Romney term in office.

    Kind of a glaring omission.

  16. KariQ says:

    @Geek, Esq.:

    As Josh Marshall noted a few weeks ago, the side that’s taking issue with poll methodologies is likely to have a depressing November 6.

    I’m not taking issue with the polls. Just the opposite. I accept the polls, all of them. You’re the one trying to narrow it down to just the ones you like, and that’s what I’m criticizing.

  17. KariQ says:

    @Geek, Esq.:

    Sorry to reply again, but I just realized you were responding to this that I said earlier:

    Gallup’s LV model – and they’ve admitted this themselves – doesn’t work well until the week before the election so relying on them is especially risky.

    I do not mean by this that Gallup’s results should be ignored. I would never suggest such a thing. Just that they shouldn’t be the only or even the primary poll that you rely on. No single poll is going to be as reliable as the aggregate of all polling.

    ABC over-sampled Democrats, I’m told, so they got a result that was too favorable to Obama. Kos is saying the PPP poll probably got too many conservatives and gave a result too favorable to Romney. Rather than playing these sorts of parse-the-numbers games, just average them all together. If the polls are truly random, then this gives you a larger number of responses to draw on and minimizes the error. That’s my point.

    But to maintain, as you did, that Romney is leading by 2-4 nationally, you need to ignore the ABC poll, the GWU poll, the SUSA poll, and the TIPP tracking poll. (Okay, the last one may be a good idea to ignore).

    To maintain, as you do, that Obama “is barely ahead” in Pennsylvania, you need to ignore virtually every poll that’s been conducted in the state.

    Who is questioning the polls again?

  18. Geek, Esq. says:


    I don’t consider online polls to be polls–they’re science experiments at this stage.

    If you look at the major pollsters:

    Pew: R+4
    Rasmussen: R+2
    Gallup: R+4
    PPP: R+4
    ABC/WaPo: O+3
    Fox News: R+1
    Politico: O+1 (with Romney ahead and more likeable in swing states)

    Note that Fox News poll, though it had Obama only behind a point, had him at 45%, which is bad for an incumbent.

    The clear outlier here is the ABC/WaPo poll (with an implausible D+9 sample). I’m not in favor of weighting polls for party ID, but when a poll does produce something with an unlikely split, that is grounds for skepticism regarding the sample quality.

    Note: I want Obama to win. I have a wife with a pre-existing condition. But, the dynamics in this race are tilting towards Romney on an increasing basis, daily. Obama’s firewall in Ohio is showing cracks–how much longer before it crumbles?

  19. KariQ says:

    The SUSA/ Monmouth poll is a phone poll. IBD/TIPP is a phone poll. If your reason for excluding them is disliking internet polls, then you are mistaken. Including the Politico/GWU poll and the TIPP tracking makes that O+3 much less of an outlier. Personally, I think it makes sense to drop the Pew poll based on it being over a week old.

    Besides, if you’re worried about ABC over-sampling Democrats, it’s worth noting that the Pew poll had more Republicans than Democrats. That would be a historical event. I think you’d have to go back to 1928 to find a presidential election in which more Republicans voted than Democrats. The same argument would go for Rasmussen; he continues to weight with a higher percentage of Republican voters than Democrats.

    Which, again, is why you shouldn’t be trying to play these games. If one poll over-samples Democrats, another is likely to over-sample Republicans. Average them all and you’re far more likely to be correct than if you start tossing out polls that don’t fit your narrative.

    Go to the page for 2008 and work with it yourself. I have and I’ve found that you get the best results by leaving in all the polls.

  20. John D'Geek says:

    PA boy here.

    I read the Susquehanna poll and, ironically enough, I think they’re probably alright. They have a solid understanding of the divisions that exist in PA, accurately point out that the far western suburb of New York City eastern PA is different than the rest of the state. Their terminology also shows me that they understand the geographic differences of PA very well (for example, they refer to the “Central/’T'” Area of PA — which makes perfect sense to me since that’s where I grew up).

    That said, the Susquehanna poll does have Obama ahead, and I don’t seriously expect Romney to pick up my state. It’s a case of “too little, too late”.

  21. KariQ says:

    @John D’Geek:

    Susquehanna’s poll was also taken in the three days following the first debate when Democratic spirits were at their lowest and Republican enthusiasm at fever pitch. I’m sure that the same poll taken by the same pollster in the same way would yield much better results for Obama if taken a week later. And after last night’s debate, it would look better still.

  22. Eric Florack says:

    Anecdotal, certainly but I’ll tell you I’m in PA several times a week… and very seldom see Obama signs as I did in 08… even in Leftist strongholds like Philly… and I’m seeing lots of Romney signs and stickers. That alone would seem a change.