What to Watch Today in Pennsylvania

In a Pennsylvania contest pitting two Democratic powerhouses, one putting together a coalition of blacks and educated white suburbanites, the other counting on the support of white ethnic, union, and blue collar voters, the former can take the state in the right circumstances. How do we know that -- because that is exactly how Fast Eddie Rendell beat Sleepy Bob Junior in the 2002 Pennsylvania Gubernatorial primary. This year, Obama is relying on the same coalition that swept Rendell to victory, while Hillary is courting the Casey coalition. Ironically, but not unexpectedly, Casey is backing Obama and Rendell is stumping for Hillary.

Rendell took only ten counties that year -- but he took them by wide enough margins to offset Casey's middle and northeastern state strength. Philadelphia went to Rendell by a couple hundred thousand votes and he swamped Casey in the elite Philly suburbs, where he won nearly 80% of the vote. He also took the Allentown area of the state by a big margin.

Unfortunately for Obama, he hasn't been able to generate Rendell-like leads in those areas of the state this year. But there are two lurking issues that Rendell and Casey didn't have to deal with which could disrupt expectations -- the invisible seven percent and the disproportionate proportional award of delegates.

The Seven Percent Solution

Three months ago, the Obama forces began a voter registration drive in Pennsylvania. It wasn't as massive as the ones they mounted in other states, but it was significant enough. By the time the deadline to register came in late March, the Democrats in Pennsylvania found themselves with over 300,000 new party affiliates. County voter registration offices were overwhelmed by the response and it is only within the last few weeks that they managed to get the newly-minted voters on the rolls.

All those polls of registered democrats in the state -- you know the ones that have been showing a steady eroding of Hillary's once-20-point-lead down to the margin of error? Well, those polls were taken from a registration list that doesn't include those 300,000+ newly registered democrats.

That group, invisible to the pollsters, could make a significant difference in the results that we see today.

Why?

Well, for one, the new registrants are much more likely to actually turn out and vote, even in this year of the large turnout. Turnout, particularly in Philadelphia and the 'burbs is of particular importance to Obama as he works to cut into Hillary's mid-state strengths. Rendell succeed in '02 by pushing record turnouts in Philadelphia and the surrounding suburbs.

For another, estimates are that the new registrants are breaking Obama at a clip well in excess of 60-40. Add to Obama's numbers 63% of the 38,000 new dems in the union and working class towns in the Lehigh Valley (Allentown/Reading), and you understand why Hillary and Bill have spent so much time there in the last few weeks. There are nearly 30,000 new dems in Pittsburgh, and almost 5,000 more in NEPA (Wilkes-Barre, Scranton) -- Hillary's Western and Eastern presumed strongholds. And in Obama's strength, Philadelphia, they added over 50,000 new Democrats.

The new democratic voters amount to about 7% of the expected turnout today -- keep that in mind when looking at the polls and their MOEs -- seven percent of today's voters were excluded from the polling pool. It might not amount to a new New Hampshire, but it could serve to explain why at the end of the day Hillary's expected ten-point-plus win winnows down substantially.

Disproportionate Delegate Scheme

In putting together the rules for this years delegate selection process, the Pennsylvania Democratic Party opted to award disproportionately more delegates to congressional districts that have gone democratic in the last two elections. This means that winning a reliably-democratic district will net more delegates than winning a similarly-sized republican district.

Not good news for Hillary as her strength comes mainly in the reliably-republican areas of the center and west states. Obama is more likely to win more of the reliably-democratic CDs than Hillary and, even if he does lose the popular vote by six, seven, or even eight points, he could walk away with more delegates -- or shave Clinton's net gain to meaninglessness.

Watch the Turnout

Rendell's victory over Casey in 2002 was based on overwhelming numbers -- huge margins in Philadelphia, 80% of the vote in the Philadelphia suburbs, and big wins in the Lehigh Valley. Polls don't show Obama with Rendell-like numbers in any of those areas.

But, a large turnout in those areas will help Obama threaten the Clinton-expectations. In Philadelphia, a large turnout early will be extremely significant as the African-American population there tends to vote later in the day. If the Philadelphia turnout is high by lunchtime, it could mean that Obama stands to post Rendell-like numbers there.

Obama could win this with high turnouts in key areas of the state. If you start to see numbers like 70% turnout in the Philly suburbs and 60% in the city, and high turnouts in the State College area (home of Penn State's main campus), and in Lancaster in South Central Pennsylvania, which has seen an influx of new residents from Philadelphia, Baltimore and DC, it could mean a surge of Obama supporters showing up to create a rough night for Hillary.

On the other side of the ledger, NE PA (Wilkes Barre, Scranton), Erie County, and Pittsburgh and surrounds, are Clinton counties and strong turnout here could favor her. (Careful here, though. Erie added almost 7,000 new democrats this year and Pittsburgh nearly 30,000.)

Wait for It

The early numbers should show a huge Obama lead -- if they don't, Obama is in for a rough night.

But don't jump to assumptions. The Central and Western counties in Pennsylvania are notoriously slow reporting and that is where Hillary is counting on to win this thing for her (apart from NEPA).

Early Night

This isn't going to be another New Mexico, though -- the results will be known relatively early, 9:30, 10 pm Eastern, at the latest.