Sunday, April 20, 2008

Defining "Victory" in Pennsylvania

If you listen to Obama supporters, Clinton has to win by a massive amount to carry on, while her supporters say a small victory is good enough. Most of that spin is largely irrelevant, all that matters is the game of expectations as it relates to the media. Overall, it would appear the media frame is Clinton must win by at least 8-10% to claim "victory"- that would be enough to give her campaign oxygen, raise further doubts about Obama and keep the superdelegates at bay. Anything less and there will probably be great pressure on Clinton to concede.

At the moment, the polling averages give Clinton about a 6% lead, which sets the stage for an interesting night. However, the two most recent polls bring some promising signs for Clinton. An American Research Group poll, finished last night puts the race at:
Clinton 54%
Obama 41%

A caution, ARG's track record has been spotty this primary season.

Speaking of "spotty", Zogby is conducting nightly rolling polls for Pennsylvania, and the results show movement for Clinton. Tonight's poll has Clinton leading by 6 points, up from 3% the night prior. What seems particularly good for Clinton, the last day of polling:
John Zogby: “A big one-day of polling for Clinton. If a 10-point victory is the pundit-driven threshold she needs on Tuesday, it looks like she can do it. This does not look like a one-day anomaly – undecideds dropped to only 5% in this latest single day of polling, and they are breaking Clinton’s way. As I suggested yesterday, if white and Catholic voters, who still are the biggest portion of undecideds, actually vote, Clinton will have her double-digit victory. Just today alone, she polled 53% to Obama’s 38%.

For Obama supporters, there is the Mason-Dixon poll, which stopped polling on Friday. That well respected outfit only gives Clinton a 5% lead, with 8% still undecided.

I would argue that an 8% margin is enough for Clinton to claim victory. While that means little in the way of a delegate dent, it still represents a solid result, in another key state, within the environment of Obama supposedly "unstoppable". I accept the lower margin, primarily because Obama has brought out all the stops in Pennsylvania, he has campaigned very hard, established an impressive ground game, not to mention outspending Clinton on ads a full 3-4 times(incredibly important in a big state).


Scott Tribe said...

As you say, Zogby hasn't been much better this primary.

Steve V said...


The only thing, and his record hasn't been good (funny because he was the darling in 2004), this is the first time he has shown a large break at the end. Previously, he had overstated most of the way, but this time, that 13% lead for Clinton comes on the eve of the primary. It is curious that his polling is moving so much, so fast. We'll see.

Steve V said...

Sorry, 15% lead for Clinton.

NA Patriot said...

Good post. One thing stood out for me and that is the notion that a Clinton victory somehow raises "further doubts about Obama". Given the demographics here, if Clinton were a viable candidate, she'd win the state by at least 20. So, if and when she wins it, by any margin, it won't raise doubts. Rather, all it will do is forestall the inevitable. Indiana is actually the big test. If she loses there, combined with a blowout in North Carolina, then we'll see some huge pressure for her to concede. Win or lose in PA, she's staying until May 6, I'd guess.

BlastFurnace said...
This comment has been removed by the author.
BlastFurnace said...

Sorry, had to delete my last comment due to a posting problem.

The real story for a lot of people is how well Obama has done overall, his gaffes nothwithstanding. He's stated a 10 point loss in Pennsylvania would still count as a win for him and he's still well positioned to do that, provided he can pull most of the remaining undecideds his way.

If the samples have lied to the pollsters, however, and Obama pulls off an upset, it's game over -- the race in Indiana would be almost irrelevant at that point. An Obama win is not likely, however; the marathon root canal will continue for at least a couple more weeks.

Steve V said...


The demographics in Penn are really quite representative of the Democratic base, latino vote aside. The reason why people initially gave Clinton such a good chance was based on her previous victories in other states, her track record. Given the way Obama has campaigned, a complete effort and the massive expenditure, the earlier 20% lead was bound to close. In retrospect, Obama may have been better off not making such a large push, analogous to Clinton in Iowa. The expectations were raised because of the focus, the ad buys, etc.

I honestly think it fair, if Clinton wins "big" for people to ask questions. I mean, when you look at the states Clinton has carried, those are one's the Dems desperately need (Penn is not a slam dunk). Given the "inevitable" factor, is it not relevant that Clinton can still win, and win decisively? All that said, North Carolina will be key no matter what.

As an aside, I haven't really watched much in the way of stump speeches the last few weeks. When I watched the debate, I was immediately struck by how tired and worn Obama looked. I didn't see the usual sparkle, he almost looked down, like he'd aged.

Steve V said...

"He's stated a 10 point loss in Pennsylvania would still count as a win for him"

Yes, and her people say a 3 point win is enough. Like I said in my post, it's a game of expectations, based on the media frame. I guarantee you this, if she wins by 10 points, nobody outside of the campaign will be claiming any victory.

Should be interesting... At least some people get to vote ;)

Anonymous said...

Noting is surprising. The polls that has her leading by less had higher undecideds, so high in fact that PPP, which iscoming out tomorrow has decided to ask for leaners, which it has never done before.

Its undecideds In Ohio and Texas were 2% on its second to last poll and 13% in Pennsylvania.

Todays polling by Zogby has Clinton up by 15%, 38-52. All the undecideds down from 11% to 5% went to Clinton. The two day average is still 6% because she polled 2 less than obama yesterday.

However, Zogby is calling a double digit win for CLinton and coming from Zogby, that's saying something.

"A big one-day of polling for Clinton. If a 10-point victory is the
pundit-driven threshold she needs on Tuesday, it looks like she can do it. This does not look like a
one-day anomaly – undecideds dropped to only 5% in this latest single day of polling, and they are
breaking Clinton’s way. As I suggested yesterday, if white and Catholic voters, who still are the
biggest portion of undecideds, actually vote, Clinton will have her double-digit victory. Just today
alone, she polled 53% to Obama’s 38%."

Anonymous said...

na Patriot,

She just might win it by 20.

Zogby is polling her at 15% with 5%undecided.

Zogby underpolled Clinton in Ohio by 10%.

He was off in NH by 16% and California by a whopping 23% all in Obama's favour.

I'm not saying its going to happen. I'm just saying if John Zogby says she 15% up that's up!

If he loses by 15% while outspending her 3:1, he may be in trouble.

(latest projections are 10 million in ad buys for Obama 3.3 million for Clinton)

Steve V said...


That quote from Zogby is in the post ;)

One caveat with all the polling will be turnout, which could bring a surprise either way.

Anonymous said...

Sorry to babble but it's not just the percentage win or loss, it's who votes for whom.

If Obama is only holding on because of getting 90% of the black vote andhe is losing the white vote 2:1 that isn't good for him.

Black voters are concentrated in blue states or red states.

if he can't win over white voters in swing states, he can't win in Nov, or at least that's the arguement that will be made by Clinton and it is pretty convicing.

If she wins Penn by 15% and Indiana by around 10%, and white voters in NC, big questions will start to be raised about his electability.

He is outspending her 5:1 in Indiana right now. At some point this starts to look bad when it doesn't have any effect or not enough effect.

By the way the last poll had her ahead in Kentucky by a whopping 36%,which is why Obama is trying to push her out of the race before WV, Kenn, and PR vote.

Anonymous said...

Sorry Steve, my jaw is just still hanging open from that Zogby quote.

For Zogby to call it for Clinton two days ahead of time by double digits is sort of gobb stopping.

As you know his polling has not been kind to Clinton this cycle. Maybe he's trying to make her look bad, if she doesn't win by that much. It's so strange I'm doing reverse pschology on it to figure out what it means.

Sure turn out is all important.

I don't know what things look like on the ground, but from the cybersphere it looks like Clinton has a massive GOTV machine in Penn and I read that Obama was not going to pay street money in Philly and black activists were upset because they saw alot of money going to white people in the campaign and for ads but Obama wasn't going to give lunch money and gas money to GOTV workers.

I bet he changed his mind on that one, but I don't know.

She has a huge machine behind her with all the election machines of Nutter and Rendell et al.

I also think there is one more thing that will work in her favour from now on. Every Clinton supporter knows that if they don't go out and vote for her she is finished. Its the NH phenomenon.

Obama supporters know that if they can't make it to the polls he is probably still the nominee.

ALl this calling for her to get out of the race and saying she needs double digits to stay alive, motivates people to vote for her, even if they are undecided.

Makes sense right. If you are truely undecided at this point, you vote Clinton, because the race goes on.

Just a thought. Anyways we'll see on Tues. Well we'll see what SUSA says tomorrow first.

Anonymous said...

Blast funace, I really don't think anyone ever lies to pollsters.

Pollster use too loose a screen for LV (all registered voters) or oversample one area (Zogby in SF, calif.) or weight based on assumptions of demographic turnout which were all wrong this primary season. Weighting also caucuses problems when you have lets say 8% black repsondents but you weight them to assume they are going to make up 20% of the vote.

No one lies though.

Steve V said...

"I also think there is one more thing that will work in her favour from now on. Every Clinton supporter knows that if they don't go out and vote for her she is finished. Its the NH phenomenon."

That's actually a really good point. Motivation, as one of the intangibles, could make a difference.

NA Patriot said...

Several folks suggest, for various reasons, that he might be "in trouble". What planet are you on? It's already over. There is no way in hell the superdelegates give this to her. Have you noticed the trend? She needs 60% of the remaining delegates. She's getting maybe 20% of the supers, and at best, might get 55% of the pledged. There is no race left. The only issue really is how she departs. I say it happens after she loses North Carolina big, and Indiana small. After a small win in PA (under 10), those two losses will have everyone calling for her to concede.

And one other thing -- there are several reasons why Obama campaigned so hard in PA: he had the money, the time (what else can he do?) AND, the popular vote matters with respect to delegates, and public perception. Long story short, Clinton needs an obvious out. IN and NC will give it to her.

Steve V said...


That's all well and good, but the fact you argue she is finished, is exactly why a big victory shouldn't happen. What does it say about the presumptive nominee, if he were to get beat convincingly by someone who has no chance? That's the problem with the argument, the more you point to the long odds, exits, the more it makes a victory "troublesome". Clinton will have won the last three major states, despite all this evidence to suggest Obama has it "wrapped up". Didn't the voters get the memo?

I've said this all along, Obama has to close the deal, if he looks like he can't, then it is fair to ask questions. What is happening now has importance beyond the math, because it needs to validate the prior. If Obama can't seem to win pivotal states, then your reason demands you ask some questions.

Steve V said...

Strategic Vision poll out tonight (18th to 20th), shows Clinton up 7:


Anonymous said...

NA patriot, with MI and Fl included the delegate count separating them is 18.

The Obama camp arguements that its impossible all hinge on not including MI and Fl.

They will be included. They were always going to be included.

They can't not be included.

There also by the way is no mechanism for anyone to arbitrarily dicatate against the Secretary of State for MI and Fl how delegates will be alloted.

The have to accept the delegates as is. They may be able to discount their votes by half, but all the Mi and FL delegates have to be accepted as determined by the Secretaries of State for FL and MI.

I honestly place bets on them being seated as is, with a full vote.

Dean and Pelosi are trying to end this before that happens, or hope that voters perceive it as over and give Obama a big enough margin that FL and MI wont matter, but if he is under 100 delegates seperating him and Clinton, they are tied.

In Michigan by the way he is claiming 40% of the votes and delegates when he only won 25% of the vote based on exit polls.

So counting Michigan as is actually allows him to steal 15% of the vote from John Edwards.

MI and FL will be seated.

Steve V said...

A pile of polls out today, seven I believe. Pretty good consensus, Clinton up 5-10% in six, with one showing Obama up 3%.

Anonymous said...

Curiouser and curiouser.

SUSA has Clinton up by 6 for the same days plus one earlier that Zogby has her up by 6, but Zogby reported as you know a big shift on the 20th.

So, I thought I'd never say this but I'll go with what Zogby says for his final poll.

If the most consistently acurrate for Clinton and the monst consistently low for Clinton agree over those days then, it's probably right.

It is not however the 6% as averaged over the 19th and 20th with undecideds breaking late.

This is possibly however the most curious polling of any primary.

The swings are wild, even in national polling.


Anonymous said...

There are still WAY to many undecideds in many of these polls, as high as 9% two days before the vote.

Very strange. It should be somewhere around 2% at this point in time, even the 5% of a couple polls which come closest is too high.

Anonymous said...

This is cross-posted at Red Tory's blog.

The PA result has led me to this conclusion. This from one who has blogged positively towards Obama. However, he has yet to close the deal.

This is not Bobby Kennedy winning the California primary in 1968, nor Jimmy Carter sealing the deal at Penn in 1976. Obama tried, but a ten point Hillary win means that there are things to play for.

Hillary won because she has now promoted herself as the Democratic version of Duncan Hunter. Ratched up the nasty metre and threatened a trade war with China and Canada.
Talked about wiping Iran off the map as a defence against Israel. She is having cross-over appeal with the Reagan Democrats.

I will look at this as a missed opportunity for Obama. The authentic "cultural disposition due to economic anxiety" neo-Marxist argument only made traction in the college towns of Penn, but not enough. If only Obama had more traction in Western PA, it may have helped.

Looking forward to your assessment.