The Phantom's "Quick" Look at the Polls II

[BluePhantom]

Well thus far, the stuff I called in my last thread is pretty much spot on. Especially this one.

Anyhow, enough of the gloating.

Where are we today? Well pretty much exactly where I said we would be and pretty much for the exact reasons why. Romney now leads in the overall RCP average, but now I have to be fair. A lot of his lead is boosted by TIPP and Pew (both of which are God awful rotten). In reality it's probably Romney +1 right now. Let's have a look at the key states:

Florida
Ignore the Marist poll. They suck and the poll was taken before the debate. I would also ignore the UNF poll. With the exception of Quinnapiac, university polls are usually ridiculously biased and the UNF poll seems to be following the same path. WeAskAmerica (WAA) I am not terribly comfortable with yet. That leaves Rasmussen with Florida at Romney +2 which means it's probably Romney +1

Colorado
A tricky one. McLaughlin and WAA are pre-debate. Ignore them. University of Denver is a college poll. Out it goes. That leaves Gravis, Rasmussen, and ARG. Mathematically between the three it's Romney +2 in Colorado. I'd say that's aggressive. Probably more like Romney +1.5

Iowa
This is critical. The only poll since the debate is Rasmussen with Obama +2. The rest are irrelevant. But this is huge because Romney must have Iowa. Even if he takes Ohio, Virginia, and Florida he needs one more state to ice it. I thought Iowa was the most realistic but based on Rasmussen he still has work to do there. Iowa's importance just shot up significantly.

Missouri
RCP has it as a swing state. I find this curious since RCP usually uses a 5.0% margin as the point where they go from "toss up" to "leaner". Only with Missouri they are not doing that even though Missouri is over that 5 point spread in Romney's favor. This piques my interest. Regardless it's a fantasy to suggest that Obama will take Missouri right now. That's all Romney's to lose.

Nevada
Still leaning narrowly toward Obama. Throw the PPP(D) poll out the window. I think we have effectively documented their biases by now. Granted it's a statistical tie but so are a lot of the others I mentioned above. Right now I still think Obama squeaks this one out but we'll see.

New Hampshire
The only poll that both started and finished after the debate is the Rasmussen poll which has NH tied. I am not buying it. I am putting New Hampshire the same as Missouri right now only on different sides...Obama's to lose

North Carolina
Obama has been dead in this state for two months. Dream on liberals

Pennsylvania
RCP moved it to toss up. See notes on New Hampshire and Missouri. This is Obama's right now.

Virginia
Toss PPP(D) out the window for reasons already stated. Forget about Marist too. They suck and their poll was pre-debate. That leaves Romney with a two point edge. Pretty much exactly what I said.

Wisconsin
Throw out Marquette as a pre-debate poll and pitch out PPP(D) as well. That leaves Obama +2 by Rasmussen. Pretty much what I expected. This state may get close but I still have it going to Obama.

Ohio
The crown jewel. Pitch out CNN...media poll...worthless. Basically what we have is a dead on tie.

So what does all this mean (except that I have thus far called this one the nose [ok maybe I am not done gloating]). Well the main thing to watch now are the number of undecided voters. The majority will break for Romney. Historically they always break for the challenger. So take a state like Ohio where Romney (on average) leads by 0.8% with 5.2% undecided. Historically 3%ish will break to Romney and that's enough to carry the state. Iowa on the other hand...if we look only at Rasmussen (since it's the only one conducted after the debate) we see Obama +2 with 4% undecided. 2.5% ought to break for Romney and that's just barely enough.

So in essence it's all Romney's ballgame right now. Tomorrow's debate may swing things a point maximum but probably not even that much. Not a bunch of ground will be gained by either side but ground can surely be lost. Look for the big gaffe that will either stop the bleeding by team Obama or open up the gushing wound even larger.

The real test will come in Obama/Romney II next week and Obama/Romney III the week after. Romney needs one more debate win to ice this. Obama isn't going to come unarmed again and the next debate is a town hall format. That favors Obama so Romney must at least pull a tie. The third debate favors Romney big because it's on foreign policy and if you are watching the news you can already see team Romney start to bring those issues to the forefront.

I don't expect another "Disgraced in Denver" Obama performance, but I do think in the third debate Romney can put this one away. Right now it's all moving in his favor.

I predict a Ryan win over Biden tomorrow (although Biden will manage to avoid a major gaffe). That will add just slight momentum toward Romney by no more than half a point at most. Obama will squeak out a narrow win In Obama/Romney II next week and that will result in stopping the Romney surge and maybe even reversing the trend by...oh...two points. Then Romney will win the third debate taking most of the undecided voters and squeaking out the presidency.

 

[BluePhantom]

An update on New Hampshire. Today ARG released a poll showing Romney +4. Now ARG is not one of my favorite pollsters. They can be a little erratic sometimes but they are far better than media, media affiliated, university, or politically affiliated polls.

I take that +4 with slight skepticism and balance it against the Rasmussen survey of a 48/48 tie. The WMUR/UNH poll I completely ignore. It's a media/university poll (bad) that was started well before last week's debate (outdated). My guess is that New Hampshire is probably Romney +1...Romney +2 right now. That being said I am not sure it will hold. That will all depend on the next two debates but Obama will have to score a win (not just a tie) in those venues to reverse what is clearly a strong Romney surge.

 

[jillian]

An update on New Hampshire. Today ARG released a poll showing Romney +4. Now ARG is not one of my favorite pollsters. They can be a little erratic sometimes but they are far better than media, media affiliated, university, or politically affiliated polls.

I take that +4 with slight skepticism and balance it against the Rasmussen survey of a 48/48 tie. The WMUR/UNH poll I completely ignore. It's a media/university poll (bad) that was started well before last week's debate (outdated). My guess is that New Hampshire is probably Romney +1...Romney +2 right now. That being said I am not sure it will hold. That will all depend on the next two debates but Obama will have to score a win (not just a tie) in those venues to reverse what is clearly a strong Romney surge.

I suspect the 'strong romney surge' stopped last night.

that said, it's now up to the president.

 

[Avorysuds]

How do you factor in Gary Johnson who is on the ballot in 48 states and all 9 battle ground states like Ohio where he polls 10%?

 

[Dr.House]

An update on New Hampshire. Today ARG released a poll showing Romney +4. Now ARG is not one of my favorite pollsters. They can be a little erratic sometimes but they are far better than media, media affiliated, university, or politically affiliated polls.

I take that +4 with slight skepticism and balance it against the Rasmussen survey of a 48/48 tie. The WMUR/UNH poll I completely ignore. It's a media/university poll (bad) that was started well before last week's debate (outdated). My guess is that New Hampshire is probably Romney +1...Romney +2 right now. That being said I am not sure it will hold. That will all depend on the next two debates but Obama will have to score a win (not just a tie) in those venues to reverse what is clearly a strong Romney surge.

I suspect the 'strong romney surge' stopped last night.

that said, it's now up to the president.

Lat night's debate will have ZERO effect on the polls...

You can pretend it will, though, if it helps you sleep at night...

 

[BluePhantom]

An update on New Hampshire. Today ARG released a poll showing Romney +4. Now ARG is not one of my favorite pollsters. They can be a little erratic sometimes but they are far better than media, media affiliated, university, or politically affiliated polls.

I take that +4 with slight skepticism and balance it against the Rasmussen survey of a 48/48 tie. The WMUR/UNH poll I completely ignore. It's a media/university poll (bad) that was started well before last week's debate (outdated). My guess is that New Hampshire is probably Romney +1...Romney +2 right now. That being said I am not sure it will hold. That will all depend on the next two debates but Obama will have to score a win (not just a tie) in those venues to reverse what is clearly a strong Romney surge.

I suspect the 'strong romney surge' stopped last night.

that said, it's now up to the president.

We'll see over the next few days but I highly doubt it. The VP debate is usually irrelevant. Regardless of which side people think won the overall effect will be basically zilch. About 29 million people watched the debate last night compared to 67 million who watched last week and the news is about Biden's obnoxious behavior for the most part. Frankly, how it's portrayed in tomorrow's SNL skit will probably have more of an impact than the debate itself (and they will surely hammer Biden's performance because that's where the comedy lies).

 

[BluePhantom]

How do you factor in Gary Johnson who is on the ballot in 48 states and all 9 battle ground states like Ohio where he polls 10%?

You are referring to the recent Gravis poll I assume. Generally I don't take one poll at face value (you should know that by now) and I would like to see more on the issue and unfortunately there are not a lot of polls that add it in specifically. It could certainly be a problem for Romney though as Johnson tends to draw more support from those who would otherwise vote for Romney.

Could Johnson hand Obama four more years? Oh absolutely! My guess is that Romney's Super PACs will start a pretty aggressive campaign of "a vote for Johnson is a vote for Obama". Whether it is effective or not remains to be seen, but certainly Romney will have to extend his current surge to compensate for the Johnson factor.

 

[Avorysuds]

How do you factor in Gary Johnson who is on the ballot in 48 states and all 9 battle ground states like Ohio where he polls 10%?

You are referring to the recent Gravis poll I assume. Generally I don't take one poll at face value (you should know that by now) and I would like to see more on the issue and unfortunately there are not a lot of polls that add it in specifically. It could certainly be a problem for Romney though as Johnson tends to draw more support from those who would otherwise vote for Romney.

Could Johnson hand Obama four more years? Oh absolutely! My guess is that Romney's Super PACs will start a pretty aggressive campaign of "a vote for Johnson is a vote for Obama". Whether it is effective or not remains to be seen, but certainly Romney will have to extend his current surge to compensate for the Johnson factor.

Fair enough. The only 2 polls I saw had GJ at 3% then I think like a month later at 6%. I don't pay much attention to polls as things can change so rapidly and the simple fact is no poll is correct seeing as the answer will only happen when votes are cast and counted, not before.

What I find interesting is when GJ is taken into account Mitt can’t possibly win. GJ is on the ballot and I wonder what polling agency’s will do if they don’t include GJ and after election day all their numbers are off by 1-5% per state. That will hurt them. The down side is if they add GJ to polls then GJ’s popularity grows as he seems relevant.

 

[BluePhantom]

How do you factor in Gary Johnson who is on the ballot in 48 states and all 9 battle ground states like Ohio where he polls 10%?

You are referring to the recent Gravis poll I assume. Generally I don't take one poll at face value (you should know that by now) and I would like to see more on the issue and unfortunately there are not a lot of polls that add it in specifically. It could certainly be a problem for Romney though as Johnson tends to draw more support from those who would otherwise vote for Romney.

Could Johnson hand Obama four more years? Oh absolutely! My guess is that Romney's Super PACs will start a pretty aggressive campaign of "a vote for Johnson is a vote for Obama". Whether it is effective or not remains to be seen, but certainly Romney will have to extend his current surge to compensate for the Johnson factor.

On the other hand if you look at yesterday's release from Gravis, Romney still leads (albeit just barely and within the margin of error) even including the Johnson factor.

Among all of the likely voters, 1,313 of the poll's participants, Romney is leading with 45.9 percent of the people saying that they'd vote for him, compared to the 45.1 percent favoring Obama. A full 9 percent said that they'd vote for another candidate. ​

Gravis Polling Results: Ohio Poll-Romney Leads With Likely Voters, Obama Ahead With Registered Voters

 

[BluePhantom]

How do you factor in Gary Johnson who is on the ballot in 48 states and all 9 battle ground states like Ohio where he polls 10%?

You are referring to the recent Gravis poll I assume. Generally I don't take one poll at face value (you should know that by now) and I would like to see more on the issue and unfortunately there are not a lot of polls that add it in specifically. It could certainly be a problem for Romney though as Johnson tends to draw more support from those who would otherwise vote for Romney.

Could Johnson hand Obama four more years? Oh absolutely! My guess is that Romney's Super PACs will start a pretty aggressive campaign of "a vote for Johnson is a vote for Obama". Whether it is effective or not remains to be seen, but certainly Romney will have to extend his current surge to compensate for the Johnson factor.

Fair enough. The only 2 polls I saw had GJ at 3% then I think like a month later at 6%. I don't pay much attention to polls as things can change so rapidly and the simple fact is no poll is correct seeing as the answer will only happen when votes are cast and counted, not before.

What I find interesting is when GJ is taken into account Mitt can’t possibly win. GJ is on the ballot and I wonder what polling agency’s will do if they don’t include GJ and after election day all their numbers are off by 1-5% per state. That will hurt them. The down side is if they add GJ to polls then GJ’s popularity grows as he seems relevant.

Well most polls will list Obama, Romney, and then "a third party candidate". To some degree that will bias the results by not including Johnson by name. How much of a swing? Hard to say because there are so few polls that actually do list him by name so there's not a whole lot of data upon which to base a comparison. My guess is maybe it would swing it by a point....maybe a point and a half, but honestly that's just a guess. I have nothing to point at statistically to support that.

But yeah...conceivably Johnson could indeed hand the election to Obama if he can make enough of an impact in even one critical state (like Ohio). It's definitely something Romney is going to have to take into account.