Pennsylvania Polls, Delegates and Predictions-Updated

by: Chris Bowers

Mon Apr 21, 2008 at 10:13

• PPP: Obama 49%--46% Clinton, 4/20
  
• Rasmussen: Clinton 49%--44% Obama, 4/20
• Survey USA: Clinton 50%--44% Obama, 4/20
  
• Zogby: Clinton 48%--42% Obama, 4/20
  
• Quinnipiac: Clinton 51%--44% Obama, 4/20
  
• Strategic Vision: Clinton 48%--41% Obama, 4/20
  
• Suffolk: Clinton 52%--42% Obama, 4/20

And here are the averages:

• Simple mean: Clinton 49.4%--43.7% Obama
  
• Non-PPP mean: Clinton 49.7%--42.8% Obama
  
• Pollster.com regression line: Clinton 49.0%--43.0% Obama

By virtually any measure, it is a 6-7% Clinton lead in Pennsylvania. The trend is slightly in her direction, and early (absentee) voting probably favors Clinton since she once led by significantly more. So, I project an 8-9% Clinton victory in Pennsylvania, which will results in a 19-16 at-large delegate breakdown, and an 11-9 PLEO delegate breakdown. CQ politics projects a 53-50 Clinton victory in the district level delegates, but I think that is pretty much a best case scenario for Obama. It seems quite possible to me that Clinton could win an extra delegate from their projections in the 2nd, the 10th, the 17th, and one of the five 5-delegate districts. That makes the best-case Clinton delegate scenario 87-71, and the best case Obama delegate scenario 83-75. I'll split the difference, and call it 85-73, with Obama winning the 2nd 7-2, the 10th splitting 2-2, Clinton winning the 17th 3-1, and Clinton taking exactly one district 4-1.

What are your predictions for tomorrow?

Update: Let me expand on the possible scenarios a bit. If Obama can keep the two-candidate vote margin under 5%, then he can force a narrow Clinton delegate victory of 81-77, which would be utterly devastating to her chances. By contrast, Clinton needs to win by 12% or more to start achieving the huge, 20 or more delegate margins that she really needs to start catching up.