Barack Obama Pennsylvania Field Office Map

More fun with maps, this time a mash-up of Pennsylvania's Congressional District boundaries and Barack Obama field office locations. The basic shape of the campaign's resource allocation is to focus on defending the four-delegate districts, spend just enough money and manpower in Western Pennsylvania to hold Clinton under 70%, and fight hardest in the Philly suburbs and Lehigh Valley.

Plausible scenarios exist for anything between Clinton +3 and Clinton +19 delegates, depending on who wins the suburbs and whether or not she can break 70% in any of the five-delegate districts. In Ohio, she only managed that margin in OH-06, where literally everything was working in Hillary Clinton's favor (no campaign stops by Obama; multiple appearances by both Bill and Hillary; Ted Strickland's endorsement mattered most as it used to be his district, etc.), and she won't have that many advantages again. But Obama probably won't win PA-01 by a 5-2 margin, and he'll probably lose one or two suburban districts, and he might not get to 37.5% in one or two of the rural districts, so Chris Bowers' Clinton +10 projection as the median scenario is probably correct.

Comments

[John]

Winning PA-01 5-2 seems possible. He'd need 64.3% there.

From what I can tell, the census data is somewhat out of date, being based on 2000 figures, but to the extent that the district has changed since then, it's probably in ways favorable to Obama - working class whites moving out, Blacks and younger, more educated whites moving in. (Although there's probably also somewhat more Hispanics and Asians, so it might be a wash)

But, using that census data, the district was 45.9% Black. Let's assume, somewhat conservatively, that Obama gets 85% of the Black vote in the district. That gets him 39.0% of the total vote, right there.

Now, on to other groups. The district was 15.0% Latino, mostly Puerto Rican. Assume Obama gets 38% of the Latino vote. That's another 5.7% of the total vote. That gets him up to 44.7% of the total.

Next, the district was 4.9% Asian, mostly recent Chinese and Vietnamese immigrants. Assume Obama gets 35%. That's another 1.7% of the total. So now he's up to 46.4% before we get to the white vote.

So we've got the remaining 34% or so of the district which is non Hispanic white. Assuming Obama gets about 45%, which seems plausible, and he gets another 15.3% of the total. That gets him up to 61.7% of the vote in the district.

That's obviously not quite what he'd need. But it's certainly right there on the frontier. It really depends on the demographics of the white vote in the district. My sense is that there's a lot of Obama type whites in the district, but that there's also still a lot of blue collar white ethnics, especially in South Philly. But I have no idea what the breakdown is.

The age breakdown in the district is notable, though, and seems pretty favorable to Obama - of the voting age population, 15.1% was 18-24, and another 41.1% was 25-44. That's 56.2% of the voting age population which was under 44, and if anything it will have gotten more pronounced since 2000.

So, basically, the first district ought to be right on the border for Obama to get that fifth delegate.

[Nicholas Beaudrot]

Yeah, I think he's in the ballpark in PA-01, but that it leans slightly against him.

[ikl]

I think that he'll get 5-2 in CD 1. Blacks are overrepresented in the Dem primary because (1) whites are more likely to be Republicans or independents even in Phily, (2) turnout is probably going to be higher among blacks than among Puerto Ricans and possibly whites (3) if many of the Asians are really recent immigrants then they were less likely to be elligible to vote and probably less likely to vote even if they are elligible. I don't know South Phily well, but my sense is that the whites may be likely to vote for Clinton. But still, Obama might get to 50% just on the black vote there which makes it pretty realistic to get to 64% even if he doesn't do so well among whites.

[ikl]

I haven't studied the state in huge detail (and don't know it well at all outside of greater Philly), but from the breakdowns that I have seen + 10 seems a bit on the optimistic side if you are an Obama supporter. Philly is only 2 CDs with 16 delegates and I'm honestly not that confident about anywhere else in the state. Even if Obama is + 8 in Phily, Clinton has 19 other CDs and the at large delegates to build a lead with.

[John]

IKL - seems plausible, mostly.

My sense is that the white people in CD-01 are a fairly mixed group. You've got working class whites in South Philly, and probably some in the part of lower Northeast in Brady's district, as well, although I'm not sure how the demographics work out there. It also has some areas which will have more Obama type whites, though - parts of the areas around St. Joe's and Temple, eastern Center City, yuppies Bellavista and Queen's Village, Northern Liberties, Fishtown, etc. Lots of gentrifying areas that will be bringing in Obama voters. And, as you say, a fair percentage of the more Clinton leaning white population isn't going to be registered as Democrat. So, yeah, I think 5-2 is totally plausible.

Going past that to the overall delegate race, you want optimism? I'll give you optimism. Here's what looks to me like the best case for Obama, assuming he can tighten up the overall margins by a decent extent (which I think is possible - Ohio was looking much closer than it ended up being for a while, and Obama seems to have been hurt by the NAFTA brouhaha and the fact that a bunch of Republicans voted for him, neither of which will happen in PA)...

1 - Obama +3
2 - Obama +5
3 - Clinton +1
4 - Clinton +1
5 - Split
6 - Split
7 - Obama +1
8 - Obama +1
9 - Clinton +1
10 - Split
11 - Clinton +1
12 - Clinton +1
13 - Obama +1
14 - Obama +1
15 - Obama +1
16 - Split
17 - Split
18 - Clinton +1
19 - Split

That actually leads to +7 Obama. That being said, it seems more likely that Clinton will pick up +2 in the 10th, and +3 in at least one, if not several, of the 5 delegate districts I gave her +1 in (CD-12 seems most likely). This would also have everything going right for Obama in the Philly suburbs and Lehigh Valley, and in Pittsburgh proper, all of which he wins in this scenario.

This isn't a plausible scenario so long as Obama remains 15 points behind, but we're still a month to go, and there's no reason, yet, to think that he can't tighten the margin.

[Nicholas Beaudrot]

+5 in PA-2 is a bit of a stretch ... he needs 72% of the vote. In GA-4, he got 74% of the vote, but GA-4 has more African-Americans. Again it's right on the bubble.

ikl: the big question mark is the four Philly suburban districts (6, 7, 8, 13). If Obama can net 3 delegates there with college educated voters, he's in good shape. But if the affluent pro-choice women vote Clinton he's in trouble. In Ohio he lost the well educated suburbs, so his overall performance will have to improve by a net 5% or so to pull it off.

[ikl]

I'll defer to John on the Philly question - I was thinking that most of the whites in that district were in south Philly. So I might be pretty off. Though my impression (possibly mistaken) or northeast is that that is probably also Clinton territory at least as far as the white folks go. My impression is that CD-2 also has University City and other places where whites are likely to support Obama. Maybe I'm wrong on that - I don't know Philly that well.

I agree that the suburban districts are the key, but think that Obama might have a hard time getting over 50% in most of them even if he is very competitive there. Obama should do well in some of the affluent areas west of Philly (the main line for example), but i'm less confident about the suburbs as a whole.

[John]

Nick - PA-02 it seems quite possible for him to get 72.3%, the magic number.

The district is about 2/3 Black. He should get about 57% of the vote from Blacks alone. The rest of the population is almost entirely white, and it's very largely dominated by the kind of whites Obama does well with - students at Penn and Temple, young people in Center City, South of South, and the Art Museum area, well educated liberal yuppies in Manayunk and Chestnut Hill, well off suburbanites in Cheltenham Township. There are some working class whites in the northwest (Roxborough, notably), and perhaps some in the part of South Philly in the district (although South Philly west of Broad is, I think, heavily Black), but Obama ought to win the White vote in CD-02 by a decent margin.

I'm considerably more confident that Obama will get 7 there than I am that he'll get 5 in CD-01.

IKL - Whites in CD-01 would tend to have a lot of South Philly, yes. But some parts of South Philly are gentrifying (Queens Village, notably). It also includes Center City and Northern Liberties, which are good areas for Obama, and some of the vicinity of St. Joe's and Temple, which should also be good for Obama. Fishtown, north of Northern Liberties is, like parts of South Philly, kind of mixed, with gentrifying types mixed in with an older white working class neighborhood, as I understand it. Most of North, Southwest and West Philly are largely Black, as are the parts of Delaware County in the district. Northeast is, as you say, good for Clinton (working class whites), but most of Northeast is in PA-13. Kensington is mostly Puerto Rican, and should be good for CLinton, as presumably also Chinatown and the Vietnamese area around Washington Avenue, to the extent that those people are voting. I think Clinton will win the white vote in PA-01, and I'm not really aware of the hard numbers in terms of demographics, but there's enough spots for Obama to do reasonably well.

Basically, in CD-01, Obama will own West Philly, North Philly, Center City, and the Delaware County parts. Clinton will do well in South Philly, Kensington, and the parts of Northeast in the district (which are limited). Getting 64% seems plausible, if not guaranteed.

In terms of the suburbs, I'm much less familiar with them. Most of Delaware County is in the 7th District, which also includes some parts of Montgomery County and southern Chester County. This area is, as I understand it, kind of mixed - both more lower middle class suburbs like Media, Upper Darby, and Lansdowne, and very wealthy suburbs like Haverford, Radnor and Swarthmore.

The Sixth District, which has six delegates and is thus less important, includes Lower Merion and other well off, more liberal Montgomery County suburbs, but also large chunks of Berks and Chester County. Chester County tends to be wealthy and Republican and semi-rural. Berks County I have no sense of - the main part of Reading is in a different district. This one should split, I think.

The Eighth District is Bucks County. Wealthy, and the outer parts are fairly rural, more liberal than Chester County and including more inner suburbs (Levittown, Bensalem). This ought to be decent Obama country.

The Thirteenth District includes most of Northeast Philly and large parts of Montgomery County. The Northeast Philly half of the district ought to be strongly for Clinton, which means to win, Obama needs to do very well in the Montgomery half. I don't have a good sense of this area, either, but I think it's fairly similar to Bucks.

[Nicholas Beaudrot]

Oh, wow, I totally was looking at the wrong district, yeah, I guess 7-2 is definitely in the cards.

Looking at the 2004 results, the suburbs go from Delaware (most Dem), Montgomery, Bucks, and Chester, in that order. I didn't realized that the Thirteenth included the Clinton parts of Philly.

So let's score 7 and 8 for Obama, and 13 for Clinton. For the pittsburgh district, I think that one's a toss-up to. And I still feel like 15 leans Clinton. So Obama is +8 in Philly, +1 in the suburbs, and either -8 or -6 in the rest of the state. So we're talking about +1 or +3, unless she hits 70% in a district or two. In the bull case he takes all the suburbs and 14 & 15, so he's +7, plus the state-level delegates. In the bear case he loses all the suburbs and Pitt and all the 5-delegate districts 4-1, it's -14 plus the state-level delegates.

[John]

Don't forget four member CD-10, in Northeastern Pennsylvania, where Clinton could easily get 64% and get 3 delegates - that's more likely than her getting 70% in any of the 5 delegate districts, I think.

CD-5 seems harder for Clinton to pick up the third delegate - that's the district that includes State College, which ought to provide a disproportionate share of that district's Democratic voters, since it's a very very Republican region. And Obama should have no real trouble preventing Clinton from getting the extra delegates in the 4 delegate south central districts 16 (Lancaster), 17 (Harrisburg), and 19 (York).