The Phantom's Polling Analysis Thread

[BluePhantom]

Weekly Breakdown for 6/5/12

Well it was a better week than last time around but unfortunately not much. We have our usual two tracking polls (Gallup and Rasmussen), a CNN/OR poll, and a late Pew Research poll that came in right at the deadline. The Pew poll is rather interesting as it is an RV poll of 2388 respondents taken over an entire month which raises my eyebrow a bit. A time frame that long is something I have never been a fan of because things can change week to week and what was going on three weeks ago can dramatically impact the results. I don’t know why they did it that way as 2388 people can be surveyed in a far shorter amount of time. Gallup and Rasmussen for example do about 500 daily. I will have my weekly fit and give Pew the “Blue Middle Finger Award” for this week as they provided a poll so spread out in its time frame that it was outdated before it was even published. Way to go Pew.
With that said let's look at the statistics:

5/30/12 - 6/5/12

Base Average
Professional Polls
Rasmussen (LV - tracking): Obama 45, Romney 46
Gallup (RV - tracking): Obama 46, Romney 45
Pew Research (RV): Obama 49, Romney 42

Professional Average: Obama 46.67, Romney 44.33; Obama +2.34

Media Polls
CNN/OR (RV): Obama 49, Romney 46

Media Average: Obama 49.00, Romney 46.00; Obama +3.00

Combined Base Average: Obama 47.25, Romney 44.75; Obama +2.50


Adjusted Average
Professional Polls
Rasmussen (LV - tracking): Obama 675, Romney 690
Gallup (RV - tracking): Obama 1403, Romney 1372.50
Pew Research (RV): Obama 1170.10, Romney 1003

Total: Obama 3248.10, Romney 3065.50

Professional Average: Obama 46.82, Romney 44.18; Obama +2.63

Media Polls
CNN/OR (RV): Obama 438.55, Romney 411.70

Total: Obama 438.55, Romney 411.70

Media Average: Obama 49.00, Romney 46.00; Obama +3.00

Combined Adjusted Average: Obama 47.07, Romney 44.39; Obama +2.67


Two Week Rolling Average

Base Average
Professional Polls

Obama 46.40, Romney 44.60; Obama +1.80

Media Polls

Obama 49.00, Romney 46.00; Obama +3.00

Combined Average

Obama 46.83, Romney 44.83; Obama +2.00

Adjusted Average
Professional Polls

Obama 46.60, Romney 44.47; Obama +2.13

Media Polls

Obama 49.00, Romney 46.00; Obama +3.00

Combined Average

Obama 46.78, Romney 44.59; Obama +2.19

In state polls I continue to be frustrated which the scarcity of polling data and the unreliability of the firms providing them. The majority of state polls this week came from PPP(D), who is under considerable scrutiny, and MSNBC, need we say more? However, as always we will plug in the data as we get it and publish the results and we can debate it later.

In Florida a PPP(D) RV poll showed Obama with a 50/46 lead over Romney. This shows almost no movement at all from PPP’s last poll of Florida for the week ending 4/17 and represents a considerable departure from the trendline. Even more surprising is PPP’s data suggesting that Obama leads Romney among Independent voters in Florida by 9 points. This could however simply reflect their survey sample of only 19% Independents (slightly over 100 respondents).

In Iowa an NBC/Marist RV poll revealed a 44/44 tie. There were no crosstabs to look at so it’s impossible for me to dig much deeper into this poll. However it would be a significant shift from PPP(D)’s poll of 5/8 showing a 10 point Obama lead in the state. Personally I don’t put a whole lot of faith in either of these agencies so….take it as you wish….

In Missouri a PPP(D) RV poll showed Obama with a 45/44 lead. Again this would represent a significant shift from previous data (little of it that there was) in Obama’s favor. Even converted to a generic LV format the results would only shift to Romney +0.76. Unfortunately there’s not a lot of data on Missouri to use as a reliable basis for comparison.

An NBC/Marist RV poll of Nevada showed a 48/46 Obama lead. We only have one other poll of Nevada which was a Rasmussen poll for the week ending 5/1 showing an 8 point Obama lead. If accurate this would represent a closing of the gap, but again I urge caution with media polls in general and NBC in particular.

A Rasmussen LV poll of Ohio revealed Obama trailing Romney by two 46/44. This is pretty much right on the trendline. An NBC/Marist RV poll of Ohio last week, however, showed a 48/42, six point Obama lead. That’s an 8 point difference in data sets from one week to the next. Personally I am not buying that and as the NBC poll last week was significantly outside the trendline and Rasmussen’s poll this week supports the trendline, it’s reasonable to assume that the NBC result was an outlier.

A Rasmussen LV poll of Virginia showed a 47/47 tie. This represents about the same results for Obama since our last set of data on Virginia from professional pollsters a month ago, but rising support for Romney. When media polls are included it represents a significant shift in Romney’s favor as the media polls of Virginia are favoring Obama by an average of about seven points above the combined average of the pros.

Finally, a Marquette University LV poll of Wisconsin showed a 51/43 Obama lead. I was very disappointed as with the recall election this week I anticipated a flurry of Wisconsin polls. Instead all we got was Marquette. Interesting to say the least. Anyhow, this shows a widening gap for Obama in Wisconsin which means either the trend has shifted since previous reports showed the gap closing, or the recent recall election has simply shaken things up and it’s a mess right now. Future polls as always will tell the tale.

 

[Foxfyre]

Thanks again for all your hard work Blue.

I was watching the Wisconsin returns come in last night and will admit that I smiled a bit when Walker won decisively despite all the media reports of how close it was going to be, it was anybody's guess who would win, etc. Exit polling was astoundingly inaccurate.

Dane County (Madison) and the University of Wisconsin are so blue that it is difficult to find even a speck of pink, much less red, there, but at one point some concern was expressed that the county could slip into the red. It didn't but the fact that any concern at all was expressed was interesting.

So the mystery remains. Wisconsin folk are apparently pretty happy with Scott Walker's reforms and how that is working out for them. So, given his piss poor fiscal perfomance, why do the numbers continue to show Obama with a substantial margin in Wisconsin? And are those numbers reliable?

 

[The T]

Welcome back Blue! Hope all is OK with you and yours?

 

[BluePhantom]

Thanks again for all your hard work Blue.

I was watching the Wisconsin returns come in last night and will admit that I smiled a bit when Walker won decisively despite all the media reports of how close it was going to be, it was anybody's guess who would win, etc. Exit polling was astoundingly inaccurate.

Dane County (Madison) and the University of Wisconsin are so blue that it is difficult to find even a speck of pink, much less red, there, but at one point some concern was expressed that the county could slip into the red. It didn't but the fact that any concern at all was expressed was interesting.

So the mystery remains. Wisconsin folk are apparently pretty happy with Scott Walker's reforms and how that is working out for them. So, given his piss poor fiscal perfomance, why do the numbers continue to show Obama with a substantial margin in Wisconsin? And are those numbers reliable?

Well that's the million dollar question. It seems funny that Walker would win in what was a pretty crushing victory for the Republicans and yet at the same time Obama would expand his lead. That doesn't seem very logical to me. Now last week we had a poll from Wisconsin Public Radio...I personally give them about as much validity as TM, and this week we have Marquette University who kind of seems to bounce around all over the place.

Honestly, I don't know what is going on in Wisconsin. I think there's been a lot of stuff happening there and we're just not getting very reliable results recently. Wisconsin is a blue state and so it's logical to assume that Obama is leading there but an 8 point lead at the same time as a major Republican victory and a gap that was previously closing seems confusing at best...frankly a little hard to believe

 

[BluePhantom]

Welcome back Blue! Hope all is OK with you and yours?

Oh I am hanging in. It's going to be dramatic for sure. Thanks for the well wishes

 

[The T]

Welcome back Blue! Hope all is OK with you and yours?

Oh I am hanging in. It's going to be dramatic for sure. Thanks for the well wishes

You betcha'.

 

[saveliberty]

Recalls are tricky. Often the electorate needs a lawbreaker before they will recall someone. Differences of political opinon are not enough.

The unions were pretty agressive as well, many people didn't voice an opinion so as to not offend the union members. The fact that the program is stabilizing Wisconsin economics is a big part too. This is probably one of the biggest political chips the Republicans pick up for November.

 

[Foxfyre]

Still there have been three recalls of governors in my recollection, and Walker is the only one to survive the process.

 

[The T]

Still there have been three recalls of governors in my recollection, and Walker is the only one to survive the process.

First one in history.

 

[saveliberty]

You can recollect the 1921 recall in North Dakota?

 

[saveliberty]

Okay let's be simplistic then. Socialist leaning governors lose recalls and Conservatives win.

 

[The T]

You can recollect the 1921 recall in North Dakota?

I looked it up. It appears that Governor Lynn Frazier was recalled.


1921 North Dakota recall - Factbites

 

[saveliberty]

Oh I agree, but Foxfyre recollects it.

 

[Foxfyre]

You can recollect the 1921 recall in North Dakota?

No, but I'm probably closer to recollecting it than you are.

Just a long time student of American history and collector of trivia here.

 

[saveliberty]

Training pants when Kennedy was shot. Just saying.

 

[Foxfyre]

Training pants when Kennedy was shot. Just saying.

Worked on Kennedy's campaign. Just sayin'.

 

[BluePhantom]

Well after a couple weeks where life got in the way (and then I got sucked into a debate on a different thread) and I have done the minimum, I spent some time this morning preparing full reports for tomorrow. I just need to wait and make sure no other polls trickle in this afternoon and I will be ready to go with the Basic Report, LV Report, Trends and Demographics Report...the whole shooting match. Today however I am going to give a breakdown on the states we are tracking, what is going on in them and how their electoral votes are currently being assigned in our analysis. So let's jump into it.

Colorado (EV: 9, Status: toss up, Trend: toward Romney)
Only in the last couple weeks have we seen some polls of Colorado coming through. For the week ending 5/29 we had the Keating (D) poll, which we pretty much discredited as trash, showing a statistical tie with an Obama lean. There have been three polls since. All have been statistical ties but are showing tighter and tighter spreads. That indicates movement toward Romney although the projection through election day remains a toss up. Romney's trendline is above the extended average and Obama's is below the extended average suggesting that the current trends should pick up speed.

Florida (EV: 29, Status: leaning Romney, Trend: toward Romney)
Florida has seen some confusing results. The media polls and PPP(D) show strong Obama leads in Florida. The rest of the professional pollsters show the exact opposite. Statistically speaking, when you add everything together it's a true toss up, but the strong Romney trend among professional pollsters has landed the state in the Leans Romney category. Like Colorado the trendlines and extended averages suggest that the momentum will increase over time and Romney will increase his lead.

Iowa (EV: 6, Status: toss up, Trend: toward Romney)
Iowa is a frustrating state to look at. Only three polls since we began, but two polls over the last two weeks. The problem is that a PPP(D) poll for the week ending 5/8 showing a ten point Obama lead was the only information we had and so that poll established the baseline. The most recent two polls show Romney leads or ties (both within the MOE). So the question becomes: "has there been a major trend toward Romney in Iowa or is the baseline fucked up because of PPP(D)'s shitty polls?" It's probably a little of both.

Missouri (EV: 10, Status: leaning Romney, Trend: toward Romney)
Missouri is another state where we don't have a great deal of information. It's only been polled recently by Rasmussen and PPP(D) and as expected they show completely different things. Still, even PPP can only muster a statistical tie on an RV poll, while Rasmussen's LV results show a state that is almost out of Obama's hands completely.

Nevada (EV: 6, Status: toss up, Trend: undetermined - maybe trending Romney?)
For such a critical state you would think there would be a lot more polling done here but there have been only two: a Rasmussen poll for the week ending 5/1 showing an 8 point Obama lead and an NBC poll last week showing a statistical tie. If we accept those results as accurate then it would indicate a pretty hard Romney trend in Nevada, but I don't put a whole lot of faith in media polls. Still, as media polls generally show an average of a four point advantage to Obama, the NBC poll showing a two point Obama lead is suggestive that the Romney trend theory may be the accurate one. Without more polling it's just impossible to tell.

New Hampshire (EV: 4, Status: Leaning Romney, Trend: undetermined)
This is probably the most frustrating state of the lot. The only poll since we started has been a PPP(D) poll showing a 12 point Obama lead. Personally, I believe that about as much as I believe pigs can fly and so we have a whole lot of nothing to look at. Currently it's listed as Leaning Romney because prior to starting this series of analyses Romney was leading in the state universally. We just don't have enough reliable information to change the state's status at this point so for now we will leave it where it is.

New Mexico (EV: 5, Status: Leaning Obama, Trend: undetermined)
See the notes for New Hampshire. It's exactly the same situation. Only PPP(D) has polled the state showing a 14 point lead for the week ending 5/1. We just have zilch for data here.

North Carolina (EV: 15, Status: Leaning Romney, Trend: heavy Romney)
There's been a lot of polling in this state and even PPP is showing Romney leads now (albeit within the MOE). The baseline was established by a SUSA poll for the week ending 5/1 showing a four point Obama lead. SUSA is pretty reliable so it's reasonable to accept that as a fairly legitimate baseline. Well things have changed, and the state is flocking to Romney like crazy. Romney's trendline is way above the extended average meaning momentum is expected to increase and the opposite is true of Obama. Unless something big happens, Obama can pretty much kiss this one goodbye.

Ohio (EV: 18, Status: toss up, Trend: toward Romney)
Things started to break hard toward Romney in Ohio somewhere between 5/8 and 5/29. Prior to that Obama held leads in all the polls in the state. For the week ending 5/15 signs of cracks in the dam began to show as Quinnapiac published a poll with a two point Obama lead. It's been all Romney in the professional polls ever since and the gap is slowly increasing. Romney's trendline is above the extended average meaning he should pick up momentum while Obama's is right on the average meaning he probably won't lose or gain much more than what he currently has. It's still a statistical tie and extending toward election day suggests it will probably remain so, but a few more Romney polls, and I may switch the status from toss up to Leans Romney.

Pennsylvania (EV: 20, Status: Leans Obama: Trend: none)
Both trendlines are flat as a board with Obama showing about an eight point lead. Even with an RV vs. LV factor worked in Obama still has a lead outside the margin of error. This is pretty much Obama's baby unless something significant happens. What's amazing is that there has been almost no movement at all in the trends. It seems that no matter what happens in the news it's completely ignored by Pennsylvania voters, at least in regards to their presidential preference.

Virginia (EV: 13, Status: toss up, Trend: slight Romney)
This is a very confusing state to look at because there is so much disparity in the results. As you might expect the media polls are showing Obama leads as is PPP(D)...shocker, I know....but Rasmussen and Purple Strategies show a far closer race. The trendlines moved big this week with Quinnapiac's poll showing a six point Obama lead. I tend to think this is an outlier though as they showed an 8 point Obama lead among Independents in Virginia which is totally wacky. Additionally both Rasmussen and Purple are LV polls while Quinnapiac is an RV poll so that may explain the difference. Right now Virginia is probably within the MOE but leaning slightly toward Obama, but what is significant is that Romney's trendline is above the extended average (again suggesting he will gain momentum over time) while Obama's is flatlined. Projected to election day it's a toss up and could go either way although the movement is toward Romney.

Wisconsin (EV: 10, Status: Leaning Obama, Trend: slight Romney)
Things were really starting to tighten up in Wisconsin right around 5/15 where we were seeing statistical ties and a strong Romney trend. Since then we have only gotten two polls. One is a WeAskAmerica poll (which is crap), and the other is a Marquette poll. Both show 5-8 point Obama leads. What is really good is that because of the recall election in Wisconsin these are all LV polls so we are getting a much more accurate picture of what is going there than in other states. Around the 15th I was tempted to move Wisconsin into the toss up category, but I wanted to see more polling to support the trend. That didn't happen and so Wisconsin will remain Obama's state to lose. However what it does show is that Wisconsin is particularly volatile and a lot of movement will happen there unlike Pennsylvania for example. The trendlines favor Romney but ever so slightly. He will need to really make a push there if he wants to take the state but the data suggests it can be taken.

 

[Foxfyre]

I live here and despite being a political junky, I can't get a good read on New Mexico either. Our local afternoon rather wishy washy conservative talk show host is convinced that Obama will carry New Mexico by a wide margin.

But we were being telephone polled something like ten times a week in the weeks leading up to the primary last Tuesday and those were so obviously skewed that there's no telling how much accurate information they were producing. And since then silence. Crickets. A complete void.

With the state being so heavily Democratic, including a mostly left leaning media, it is unlikely that we will get much information other than what they want us to believe.

What I think might be a bellwether indicator will be watching the Heather Wilson (R) vs Martin Heinrich (D) senate race. If Heather makes a good showing, that could be ominous for Barack Obama. If Henrich polls well, Obama will probably carry the state.

 

[BluePhantom]

I live here and despite being a political junky, I can't get a good read on New Mexico either. Our local afternoon rather wishy washy conservative talk show host is convinced that Obama will carry New Mexico by a wide margin.

But we were being telephone polled something like ten times a week in the weeks leading up to the primary last Tuesday and those were so obviously skewed that there's no telling how much accurate information they were producing. And since then silence. Crickets. A complete void.

With the state being so heavily Democratic, including a mostly left leaning media, it is unlikely that we will get much information other than what they want us to believe.

What I think might be a bellwether indicator will be watching the Heather Wilson (R) vs Martin Heinrich (D) senate race. If Heather makes a good showing, that could be ominous for Barack Obama. If Henrich polls well, Obama will probably carry the state.

Good idea. I will try and get a feel for that race and see if we can divine anything from it. My brother lives there as well. My gut feeling is that it will be pretty tough for Romney to take NM

 

[saveliberty]

If your in the toilet, whether it flushes or not doesn't really make a difference. Gas prices have shot back up this week, the market is still down and no other economic indicator is going up. The key for Romney is to NOT give Obama ammunition. Meanwhile, we have administration officials runnning into cars and continuing to shoot themselves in the foot with the economy. Couldn't write a better script.