First Pennsylvania Results Thread

by: Chris Bowers

Tue Apr 22, 2008 at 20:00


14% Reporting
Clinton: 53% (61 delegates)
Obama: 47% (38 delegates)

Polls have closed. The first exit poll can be found here. Quick multiplication of the gender crosstabs produces Clinton 52%--48% Obama. The first exits are usually adjusted, however. I can already see a problem where the exits claim that Philadelphia and its suburbs where only 29% of the electorate. Yeah, right.

Update: One of the reasons I am focusing on the regional exit polls is because it is actually the only part of an exit poll that can be matched against actual results. And, even though right now the exit polls only show Philadelphia and the suburbs at 32% of the total vote, they will end up way, way higher than that figure. A second reason is that if Obama won Philadelphia with 70%, and the suburbs with 59%, then he won the state. A shocking victory, no doubt, but if those Philly and Philly suburbs numbers are right, then he won.

Update 2: CBS declares Clinton the winner.  MSNBC too.

Update 3: Assuming these early calls are correct, it appears that Obama's margin in Philadelphia is lower than expected. I was still right about the size of Philadelphia in the electorate, as 440,000 votes will come from there (or so), making it over 20% of the vote. However, right now Obama only leads by 10% in Philly. So, both the size of Philadelphia and the margin in Philadelphia were way wrong in the exit polls.  

Chris Bowers :: First Pennsylvania Results Thread

Tags: , , (All Tags)
Print Friendly View Send As Email
Looks good (0.00 / 0)
Those gender crosstabs are usually pretty accurate, within 4 points or so. Looks like we won't see a double-digit Clinton victory tonight.

I was very surprised that it wasn't called at 8pm. Then again, Ohio was initially too close to call, too.


How close to exit polls generally reflect the final results? (4.00 / 1)
Didn't Obama do quite well in the Ohio Exit Polls but lost by 9-10 margin by the end?

Do you guys think this will stay in margin between 3-6 points? That would be a great outcome for Obama considering the circumstances.  


[ Parent ]
Well... (0.00 / 0)
The leaked (often fake) exit polls aren't worth the pixels they're printed on. The gender/income/race crosstabs that are officially released by the networks are much more accurate.

Here's a good look from a diary at Daily Kos. The gender crosstabs were dead on in Ohio.


[ Parent ]
See Chris's earlier post (0.00 / 0)
Linking to an article saying Obama has been underperforming his exits by 7 points. Still, that might but it within 10.

[ Parent ]
calling it would ruin (0.00 / 0)
the very finely sculpted suspense :)

[ Parent ]
In Ohio Clinton won 50-48 among men, 57-41 among women (0.00 / 0)
In PA Obama takes men 53-47, Hillary takes women 55-44. Interesting.

Who should early returns favor? (0.00 / 0)
Anyone have any idea whatsoever where the early returns will come from?

Will we see results from Obama-areas or Clinton-areas initially?

Is this likely to be a long night?


Well (0.00 / 0)
Politico says the cities and suburbs generally will report first. Therefore, we should see a small lead for Obama until rural districts report in the middle of the night.  

[ Parent ]
I have to defer to a PA expert (0.00 / 0)
But if, as Chris says, Philadelphia and its suburbs are significantly more than 29% of the vote, and Obama won the city with 69%, and the suburbs with 62%, what does that mean for the overall state picture and the delegate picture?

I ask because (4.00 / 1)
the last SUSA poll had southeastern PA at 43% of the vote, with Obama winning the region 55-41.  If SUSA sampled correctly and it is closer to 43% than 29%, and CNN got the exit percentages right, Clinton might be in trouble, because SUSA only gave her a 6 point statewide lead, 50-44.

[ Parent ]
I get the same numbers (0.00 / 0)


[ Parent ]
Unless, Of Course (4.00 / 5)
All those figures are base-12, in which case all bets are off.

"You know what they say -- those of us who fail history... doomed to repeat it in summer school." -- Buffy The Vampire Slayer, Season 6, Episode 3

[ Parent ]
Ha (0.00 / 0)
that will typically do it.

[ Parent ]
and the paperless machines (0.00 / 0)
could well be calculating in base 12.

Okay, you made me laugh, Paul!


[ Parent ]
CNN has Clinton only carrying white men by 53-46 (0.00 / 0)
Slightly better than the 55-45 we heard earlier.

Based on the average of exit polls being wrong by 7 (0.00 / 0)
This would mean, final results, Clinton will end up by 11.

I expected 10 to be the high end, but really 8-12 isn't out of bounds.


Demographics suggest less than that (0.00 / 0)
In exit polls those numbers generally hold up better than the topline numbers.

[ Parent ]
Early exit polls overstate him by 7 but (0.00 / 0)
gender crosstabs released at the top of the hour are generally spot on
They were in OH

[ Parent ]
how would we know? (0.00 / 0)
Don't they just take the demographic numbers and adjust them to get the final results?  I don't think there is an independent measurement.

New Jersey politics at Blue Jersey.

[ Parent ]
Pattern is holding (0.00 / 0)
Looks like

[ Parent ]
I could be wrong... (0.00 / 0)
But I think that only goes for initial, leaked exits from the afternoon.  Exits from the after the closing of polls are generally accurate within 1-3 points and don't show a bias to one side or another.  Or so I remember from a post of Poblano's.  

[ Parent ]
What??? (4.00 / 1)
But don't the people of Pennsylvania know that Obama's pastor is angry, and Obama would rather drink orange juice than whiskey, and he thinks people whose lives have been shattered cling to things like religion?

If the exits polls are (0.00 / 0)
correct Obama won the delegate count by at least 6

cnn (4.00 / 1)
CNN and MSNBC are now saying "too early to call", that was changed from "too close to call". What does this mean?

It means... (0.00 / 0)
They're almost certain about who will win, but they're not going to say until more results come in.

IIRC, Wisconsin was "too early to call" for about half an hour before they called it for Obama.

It's good news for Clinton. But at least they haven't called it outright yet.  


[ Parent ]
I dont think (0.00 / 0)
the fact she is going to win is a shock. I dont care about the call I care about the margin. And for the record if the exit poll demographics there goes most of her rationale.

[ Parent ]
it means (0.00 / 0)
Philli is looking very bad for Obama

[ Parent ]
Clinton wins (0.00 / 0)
MSNBC just called it. Now the question is the margin.

MSNBC is calling it for Clinton (0.00 / 0)
No real surprise. The question is now one of margin.

It looks like a genuinely close race. (0.00 / 0)
If Obama is running 10 points ahead of expectations in the southeast, neither candidate can win by more than a few points.

Would the race keep going if Clinton won by one point?


Fox called it for Clinton too (0.00 / 0)
Does this mean the margin is significant?

Hard to say (0.00 / 0)
I think there is a margin (ie. more than 1-2 points) but I don't think it means the margin is necessarily 10+. I'd say they could call it this early and it could still be a 3-5 point race.

[ Parent ]
What was the margin for Philly? (0.00 / 0)
However, right now Obama only leads by 10% in Philly. So, both the size of Philadelphia and the margin in Philadelphia were way wrong in the exit polls.  

What was the margin for Philly in the exit polls?


70-30 (0.00 / 0)
Seems like an awfully strange disconnect.  

[ Parent ]
That said (0.00 / 0)
we have 25% of precincts. Not 25% of the votes. So maybe the less populated areas reported first and the high turnout AA strongholds later

[ Parent ]
No kidding. (4.00 / 1)
   Obama is under-performing pretty badly in Philadelphia.

John McCain lets lobbyists shape his economic policy

Now 57-43 Obama per CNN (4.00 / 1)


[ Parent ]
58-42 (4.00 / 1)
Looks like the AA precints were outstanding. I expect the gap to continue to grow.

[ Parent ]
60-40 now (0.00 / 0)
Looks like Obama's gaining in Philadelphia. Looks like he's going to keep it in single digits state-wide.  

[ Parent ]
probably a dumb question, but (0.00 / 0)
could someone tell me what the difference between precincts and districts is?

Right now, CNN is showing Clinton with 248K, Obama 220K with 22% of precincts reporting. The PA gov election site is showing Clinton with 278K, Obama with 253K, but only 20% of districts reporting.


Obama won Pittsburgh/PA-14. (0.00 / 0)
As I expected.  

Link here


Margin of victory (0.00 / 0)
By now, we have all heard repeatedly about the 10 point victory for HRC.  

Permit me to point out that the difference in the popular vote between the 2 candidates is 9.3%.  I was always taught the 9.3% would round to 9%.  Without considering fractions of a percentage, the winning victory was 9% - not 10% (i.e. HRC had a single digit victory).  

Of course, no one is reporting it this way or even achknowledging the point.  



Donate to Open Left








Friends of the Earth thanks the OpenLeft community for the ideas you generate and your contributions to the progressive movement.

As an anti-spam measure, there is a 24-hour waiting period after registering before new users can comment.
blog advertising is good for you
blog advertising is good for you
SEARCH

   

Advanced Search