The Phantom's Polling Analysis Thread

[sfcalifornia]

The present state of the race can be summed up as follows:

Currently, President Obama is leading in Florida, Pennsylvania, Ohio, Michigan, and Wisconsin.

Until that changes, Obama can't lose.

Is he? Well lets have a look at that. Hell let's use RCP's numbers alone and ignore mine for the moment and give a narrow margin of error of 3.5%.

Florida: Obama +0.5%. That's inside the MOE. That's a statistical tie. (BTW my spreadsheet has Obama +1 - also a statistical tie)

Pennsylvania: Obama +7. Yep...that's outside the MOE. He's leading there. (Interestingly I have Obama +8)

Ohio: Obama +4.6...outside the MOE (barely) Yep he's leading there. (wow I have Obama by 4....holy shit)

Michigan is obvious....no one even considers them a swing state. I don't even have them on my spreadsheet.

Wisconsin: Obama +1.6...well that's inside the MOE. That's a statistical tie (BTW I have Obama by 2.5)

Seems my calculations are actually giving Obama a better chance in those states than RCP's. Unfuckingreal. But to the point of your post....no he's not leading in all those states. He's leading in PA, MI, and barely in Ohio. He is tied in all the rest.


Now let's look at the trends since March


Wisconsin
Marquette: 4/29 - Obama +9; 5/12 - tie; 9 point Romney swing
Daily Kos/PPP: 4/15 - Obama +6; 5/13 - Obama +1; 5 point Romney swing
Rasmussen: 3/27 - Obama +11; 5/9 - Obama +4; 7 point Romney swing

Looks like things are going in the wrong direction for Obama in Wisconsin, huh?

Ohio
Quinnipiac: 3/26 - Obama +6; 5/1 - Obama +2; 5/7 - Obama +1; 5 point Romney swing
Rasmussen: 3/26 - Obama +8; 4/18 - Obama +4; 4 point Romney swing

Hmmm...going the wrong way there too.

Florida
Quinnipiac: 3/26 - Obama +7; 5/1 - Romney +1; 8 point Romney swing
Rasmussen: 3/13 - Obama +3; 4/25 - Romney +1; 4 point Romney swing

Looks about the same in Florida too huh?

Yeah I'll bet Obama is feeling real secure right now.

I don't think either candidate is feeling too secure at the moment.

Romney can have Florida. If Obama picks up Ohio and New Hampshire, all he needs is either Iowa OR Colorado OR Wisconsin.

 

[Foxfyre]

The present state of the race can be summed up as follows:

Currently, President Obama is leading in Florida, Pennsylvania, Ohio, Michigan, and Wisconsin.

Until that changes, Obama can't lose.

Is he? Well lets have a look at that. Hell let's use RCP's numbers alone and ignore mine for the moment and give a narrow margin of error of 3.5%.

Florida: Obama +0.5%. That's inside the MOE. That's a statistical tie. (BTW my spreadsheet has Obama +1 - also a statistical tie)

Pennsylvania: Obama +7. Yep...that's outside the MOE. He's leading there. (Interestingly I have Obama +8)

Ohio: Obama +4.6...outside the MOE (barely) Yep he's leading there. (wow I have Obama by 4....holy shit)

Michigan is obvious....no one even considers them a swing state. I don't even have them on my spreadsheet.

Wisconsin: Obama +1.6...well that's inside the MOE. That's a statistical tie (BTW I have Obama by 2.5)

Seems my calculations are actually giving Obama a better chance in those states than RCP's. Unfuckingreal. But to the point of your post....no he's not leading in all those states. He's leading in PA, MI, and barely in Ohio. He is tied in all the rest.


Now let's look at the trends since March


Wisconsin
Marquette: 4/29 - Obama +9; 5/12 - tie; 9 point Romney swing
Daily Kos/PPP: 4/15 - Obama +6; 5/13 - Obama +1; 5 point Romney swing
Rasmussen: 3/27 - Obama +11; 5/9 - Obama +4; 7 point Romney swing

Looks like things are going in the wrong direction for Obama in Wisconsin, huh?

Ohio
Quinnipiac: 3/26 - Obama +6; 5/1 - Obama +2; 5/7 - Obama +1; 5 point Romney swing
Rasmussen: 3/26 - Obama +8; 4/18 - Obama +4; 4 point Romney swing

Hmmm...going the wrong way there too.

Florida
Quinnipiac: 3/26 - Obama +7; 5/1 - Romney +1; 8 point Romney swing
Rasmussen: 3/13 - Obama +3; 4/25 - Romney +1; 4 point Romney swing

Looks about the same in Florida too huh?

Yeah I'll bet Obama is feeling real secure right now.

I don't think either candidate is feeling too secure at the moment.

Romney can have Florida. If Obama picks up Ohio and New Hampshire, all he needs is either Iowa OR Colorado OR Wisconsin.

Unless Obama can successfully pull off an impressive October surprise, which his campaign is almost certain to be looking for a way to do, this election will likely hinge on the misery index. If unemployment--the REAL numbers, not the 'posted for public consumption' ones--remains high, if gasoline prices continue to push inflationary prices onto just about everything we buy, despite the posted 'official' inflationary index, if homeowners continue to see their property decline in value and remain at risk for foreclosure because they can't make their house payments, and if Romney makes no serious gaffes or useful (for the other side) comments, the Republicans will likely take the White House come November.

If things improve dramatically and the Democrats continue to be sucessful in painting Romeny as less than a compelling figure, then Obama wins.

And none of that makes the polling data any less interesting for those of us who are history buffs from back when to history in the making now.

 

[BluePhantom]

I don't think either candidate is feeling too secure at the moment.

Romney can have Florida. If Obama picks up Ohio and New Hampshire, all he needs is either Iowa OR Colorado OR Wisconsin.

I would definitely agree. I mean it depends on what you are giving Obama to start with right now but yeah Obama getting those two states would damn near ice it. The PPP poll in New Hampshire is certainly concerning...assuming you accept their data which as I have said I am strongly questioning at this point. I think in the end Romney will choose Portman as his running mate because it virtually locks Ohio in the red and at that point the tables completely turn.

 

[BluePhantom]

Just a note on this Daily Kos/PPP(D) poll of Wisconsin. This is very very very very interesting. DK fired Research2000 as their polling agency in 2010 after a disastrous showing of inaccuracy by R2K. DK's political biases are quite apparent and as they released poll after poll showing heavy Democratic leads (which turned out to be completely wrong) they were facing heavy criticism of intentional wrongdoing. So out went R2K and in came PPP. What is so interesting about this is that regardless of any opinion you may (or may not) have about DK's website biases, it stands to reason that they want accurate polling information else they would not have fired R2K.

With that rationale it seems amazing to me that the polls PPP has released that were funded by the DNC have been showing massive Obama leads, but suddenly they release a poll of Wisconsin commissioned by DK and it's right there with other pollsters. Well Sacrebleu!! Now before I go off half cocked and start making accusations I want to see if this continues to happen. If we see a pattern where the PPP polls commissioned by the DNC are showing huge Obama leads and the PPP polls commissioned by other organizations are in line with everyone else, there can only be one conclusion and I don't think I have to spell it out for you.

The Wisconsin poll results certainly suggest something like that may be happening, but again I want to see the same thing happen over several polls before I start making strong accusations of intentional tampering. For NOW all I will say is "hmmm....this looks fishy", but I am going to be watching this situation like a fucking hawk. I will report the results as they come.

I just want to follow up on this point a little. I did some poking around and this is the second time PPP has released a poll of Wisconsin that was commissioned by the Daily Kos. The previous poll was on 4/15. It was an LV poll showing Obama+6. Now the only other poll taken in that same time frame was a Marquette RV poll showing Obama +9. So if we adjust for the LV factor that puts those two polls right about the same.

So this is the second time when PPP released a poll commissioned by someone other than the DNC and the numbers are supported by everyone else...but when it's commissioned by the DNC they are giving about 5 more points to Obama on average than everyone else. I am giving them the "three strikes rule" myself although with now two polls showing the same pattern, I have to concede it's starting to become a "pattern".

Regardless of how I decide (or you decide) they will still be included in the weekly calculations and everyone can judge for themselves whether to pay attention to them or not.

 

[BluePhantom]

A respectful request to NYCarbineer, we are having a perfectly civil discussion here on the most current polling data available and what that probably means in relation to what we know of previous races and within the current dynamics.

If you have a problem with BluePantom, please take it to the Flame Zone or someplace else, but please don't mess up what the rest of us are enjoying here. BP has done a yeoman's job of compiling and posting the figures and he has adequately and competently sourced every one. You can contribute to the thread by showing compelling data with links that would challenge what he posts, or your personal sense of what the numbers mean, or your perception of polls in general, but school yard taunts are not appropriate here.

I put him back on ignore. I am not going to play any further role in allowing him to derail the thread and turn it into a pissing match. He's just trying to piss me off.

Thanks for the comments though and helping to reinforce the spirit of the thread. I agree 100%

 

[BluePhantom]

Unless Obama can successfully pull off an impressive October surprise, which his campaign is almost certain to be looking for a way to do, this election will likely hinge on the misery index. If unemployment--the REAL numbers, not the 'posted for public consumption' ones--remains high, if gasoline prices continue to push inflationary prices onto just about everything we buy, despite the posted 'official' inflationary index, if homeowners continue to see their property decline in value and remain at risk for foreclosure because they can't make their house payments, and if Romney makes no serious gaffes or useful (for the other side) comments, the Republicans will likely take the White House come November.

If things improve dramatically and the Democrats continue to be sucessful in painting Romeny as less than a compelling figure, then Obama wins.

And none of that makes the polling data any less interesting for those of us who are history buffs from back when to history in the making now.

Yeah this pretty much nails it on the nose. It's all about the economy. People can raise hell about this side issue, and that social issue and blah, blah, blah but people vote with their wallets first. I was recently in New Orleans and was chatting with a cab driver (who was black) and he said it best: "I love Obama, I am thrilled that there's a black man in the Oval Office, and I respect him as a person....but I won't go broke for him."

Perhaps, as I am a white guy, he may have been sucking up to get a good tip...who knows, but my guess is there are going to be a LOT of moderates and Independents who feel the same way and they are the ones who decide elections.

 

[BluePhantom]

5/15/2012 Demographics and Trends Analysis


Well this will be a quick analysis this week because there just isn’t anything to really delve into. The crosstab data for this week was absolute crap and all I could get was information on ideology from two sources: CBS/NYT and the Washington Post.

Now CBS/NYT has Verifiable Data which as you may recall is when they provide the number of people in the sample who fit a demographic we are looking at and crosstabs on how those demographics answered. For the Washington Times they did not provide how many Democrats, Republicans, and Independents were in the sample so I compared it against a standard generic LV model of 32/32/36. As such the data from the Washington Times is Estimated Data.

So what we can find out this week is that when the two are averaged together, according to Ideology, Obama’s support has dropped three points among Republicans from 7% to 4%, two points among Democrats from 86% to 84%, and two points among the critical Independents going from 37% to 35%. Romney on the other hand has picked up 3 points among Republicans from 87% to 90%, one point among Democrats from 8% to 9%, but he has dropped 6 points among Independents from 46% to 40%. When you look at previous weeks what you see is an increasing number of undecided voters among the Independent voting block.

Independent Undecided Trend
4/17 = 13%
4/24 = 13%
5/1 = 21%
5/8 = 17%
5/15 = 25%

At the beginning of our calculations those undecided Independents were moving away from Obama. They still are as his number continue to fall on a pretty steady clip. Initially Romney’s numbers held strong at around 45% - 46% but his numbers are showing some movement now too on a declining trend among Independents.

Now we have to be careful here because as you know the amount of demographic data we have received on a week to week basis has been frustratingly thin and some of Romney’s decline has to be attributed to the Washington Times’ poor Romney showing of merely 38% support compared to CBS/NYT’s 43% support for Romney again among Independents. It could also be that last week’s data from TIPP could have been so whacked out that it skewed the numbers (they did show significantly stronger Obama support among Independents than the trendline suggested. This week’s data falls more in line with previous trendlines before the TIPP poll, so who knows.

One thing we can say though is that when you look at the numbers over the last 3-4 weeks and take everything as a whole, both Obama and Romney are losing support among Independents although Obama is losing it faster. Assuming things stay as they have been going, if you calculate a rolling average based on current data and extend it through election day and plot a trendline, Romney is on pace to win among Independent voters by about 9 points.

Verifiable Ideology
CBS/NYT: (among Independents) Obama 36%, Romney 43%; Romney +7

Estimated Ideology
Washington Times: (among Independents rounded) Obama 34%, Romney 38%; Romney +4

Base Average Ideology: Obama 35%, Romney 40%; Romney +5

Adjusted Average ideology: (adjusted for sample size): Obama 474.32, Romney 545.66; Obama 34.83%, Romney 40.06%; Romney +5.24

There was not enough data to calculate demographic trends for any other category this week.



In state trends

Florida had only one poll last week (and none the week before that) . The Suffolk/Channel 7 News poll showed Obama with a one point lead 46/45. Two weeks ago Romney held a one point lead in an average of Rasmussen and Quinnipiac polls taken that week (both showed a one point Romney lead). Is it a trend or statistical noise? Given that it’s only one poll and it’s within the MOE of previous data and the trendline, my guess is that there hasn’t been much movement statistically speaking. Romney’s trendline is flat as a board in Florida at 45% while Obama’s trendline continues to decline. Projected through election day according to current trends it’s a dead heat but Obama’s trendline is well below the extended rolling average meaning his decline is picking up speed overall and his support on election day is currently expected to be below the extended average. In other words Florida is presently looking good for Romney but it's far from "in the bag".

Quinnipiac polled Ohio this week and showed a 45/44 Obama advantage. According to Quinnipiac alone this shows a statistical tie just as they did a couple weeks ago. This again suggests strongly that the PPP(D) poll of Ohio last week that showed a 7 point Obama advantage is probably an outlier as more current data reinforces the average and the trendlines while the PPP poll is way outside it. What we do see is increasing support for both candidates in Ohio although Romney’s support is increasing at a faster clip. Projected through election day we have a statistical dead heat although at current paces it’s still Obama’s state to lose.

There was not enough data to look for trends in any other state. We got some good polling in Wisconsin and a PPP(D) New Hampshire but there was no previous data by which to make a comparison and identify a trend.

 

[NYcarbineer]

A respectful request to NYCarbineer, we are having a perfectly civil discussion here on the most current polling data available and what that probably means in relation to what we know of previous races and within the current dynamics.

If you have a problem with BluePantom, please take it to the Flame Zone or someplace else, but please don't mess up what the rest of us are enjoying here. BP has done a yeoman's job of compiling and posting the figures and he has adequately and competently sourced every one. You can contribute to the thread by showing compelling data with links that would challenge what he posts, or your personal sense of what the numbers mean, or your perception of polls in general, but school yard taunts are not appropriate here.

He's a rightwing propagandist who lied OUTRIGHT. I posted the facts detailing his LIE.

Since when is calling out a proven liar flaming?

 

[NYcarbineer]

Florida, Ohio, and Wisconsin are a toss up...

Stop lying...

RealClearPolitics - 2012 Election Maps - Electoral Map

He's leading in the polls. Look at the averages.

You do know what a "toss up" is, don't you?

Never mind... You're an idiot....

Do you know what it means when someone gets a higher percentage in the poll than his opponent does?

 

[NYcarbineer]

Dude I am not going to let you start shit again. Now I will respond to this once and that's it. If you want to attempt to derail a thread go do it someplace else or I will put your ass right back on ignore where you belong.

My error on the previous date of September 29. From September 5 through September 11th the majority of the polls showed McCain ahead or tied at the very least. Lehman Brothers declared they were fucked on September 9th (not 29th) and their stock dropped 45% in a day, the Dow fell 300 points that day on that news. On the 10th they announced they would be forced to sell off and all hell broke loose.

Prior to that announcement here's what the polls showed:

GWU/Battleground 9/7 - 9/11 1000 LV 3.1 44 48 McCain +4
Associated Press/GfK 9/5 - 9/10 812 LV -- 44 48 McCain +4
FOX News 9/8 - 9/9 900 RV 3.0 42 45 McCain +3
Gallup Tracking 9/7 - 9/9 2714 RV 2.0 43 48 McCain +5
Rasmussen Tracking 9/7 - 9/9 3000 LV 2.0 48 47 Obama +1
Hotline/FD Tracking 9/7 - 9/9 902 RV 3.2 45 45 Tie
NBC News/Wall St. Jrnl 9/6 - 9/8 860 RV 3.3 46 45 Obama +1
ABC News/Wash Post 9/5 - 9/7 LV 3.0 47 49 McCain +2
USA Today/Gallup 9/5 - 9/7 823 LV 4.0 44 54 McCain +10
CBS News* 9/5 - 9/7 655 RV 4.0 44 46 McCain +2
CNN/OpinionResearch 9/5 - 9/7 942 RV 3.0 48 48 Tie

Now immediately afterwards guess what...it started to turn Obama's way:

Pew Research 9/9 - 9/14 2307 LV -- 46 46 Tie
Reuters/Zogby 9/11 - 9/13 1008 LV 3.1 47 45 Obama +2
Newsweek 9/10 - 9/11 1038 RV 3.8 46 46 Tie


On the 15th (the day they officially filed for bankruptcy) the Dow dropped 500 points and McCain was completely hosed and within one week we had gone from a McCain lead to an unbelievable Obama surge:

FOX News 9/22 - 9/23 900 RV 3.0 45 39 Obama +6
Marist 9/22 - 9/23 689 LV 4.0 49 44 Obama +5
NBC News/Wall St. Jrnl 9/19 - 9/22 1085 RV 3.0 48 46 Obama +2
ABC News/Wash Post 9/19 - 9/22 780 LV 3.5 52 43 Obama +9
LA Times/Bloomberg 9/19 - 9/22 838 LV -- 49 45 Obama +4
Ipsos/McClatchy 9/18 - 9/22 923 RV 3.2 44 43 Obama +1
CNN/Opinion Research 9/19 - 9/21 697 LV 3.0 51 47 Obama +4
F&M/Hearst-Argyle 9/15 - 9/21 1138 LV 3.0 45 47 McCain +2
Gallup Tracking 9/15 - 9/17 2815 RV 2.0 48 44 Obama +4
Rasmussen Tracking 9/15 - 9/17 3000 LV 2.0 48 48 Tie
Hotline/FD Tracking 9/15 - 9/17 912 RV 3.2 46 42 Obama +4
GWU/Battleground 9/11 - 9/18 800 LV 3.0 47 47 Tie
CBS News/NY Times 9/12 - 9/16 LV 3.0 49 44 Obama +5
Quinnipiac 9/11 - 9/16 987 LV 3.1 49 45 Obama +4



Now if you don't liek the numbers I have, feel fucking free to calculate them yourself, and go try and find one fucking poll that I have ignored or misrepresented in the official reports. In short....go fuck yourself.



http://www.nytimes.com/2008/09/11/business/11lehman.html?_r=1&hp&oref=slogin

Business & Technology | Lehman shares slip on plans to auction off unit, consider sale of company | Seattle Times Newspaper

JPMorgan Gave Lehman $138 Billion After Bankruptcy (Update3) - Bloomberg

RealClearPolitics - Election 2008 - General Election: McCain vs. Obama

http:http://www.nola.com/business/index.ssf/2008/09/dow_plunges_nearly_300_points.html//

You're a liar. That's all there is to it.

The conservatives around here will praise you only because you lie in a way that makes conservatives look good.

I give you five links worth of documentation and all you can come back with is "You're a liar and that's all there is to it". Geez....I just don't know how I am going to live with this schooling I am receiving.

I gave you a list of a dozen pollsters who had Obama ahead just prior to September 29th and you ignored it.

 

[NYcarbineer]

A respectful request to NYCarbineer, we are having a perfectly civil discussion here on the most current polling data available and what that probably means in relation to what we know of previous races and within the current dynamics.

If you have a problem with BluePantom, please take it to the Flame Zone or someplace else, but please don't mess up what the rest of us are enjoying here. BP has done a yeoman's job of compiling and posting the figures and he has adequately and competently sourced every one. You can contribute to the thread by showing compelling data with links that would challenge what he posts, or your personal sense of what the numbers mean, or your perception of polls in general, but school yard taunts are not appropriate here.

I put him back on ignore. I am not going to play any further role in allowing him to derail the thread and turn it into a pissing match. He's just trying to piss me off.

Thanks for the comments though and helping to reinforce the spirit of the thread. I agree 100%

Proving you wrong and proving you dishonest should piss you off. And embarass you.

 

[NYcarbineer]

A respectful request to NYCarbineer, we are having a perfectly civil discussion here on the most current polling data available and what that probably means in relation to what we know of previous races and within the current dynamics.

If you have a problem with BluePantom, please take it to the Flame Zone or someplace else, but please don't mess up what the rest of us are enjoying here. BP has done a yeoman's job of compiling and posting the figures and he has adequately and competently sourced every one. You can contribute to the thread by showing compelling data with links that would challenge what he posts, or your personal sense of what the numbers mean, or your perception of polls in general, but school yard taunts are not appropriate here.

We have closed threads now here at USMB? And you're a moderator?

 

[BluePhantom]

Regarding Wisconsin: Marquette released an LV poll today resulting in a 46/46 tie. This supports the Daily Kos poll of Obama +1 and reinforces the trend Rasmussen suggested at the beginning of last week of Obama +4. Over the last four weeks that means Obama's lead in Wisconsin according to Overall Base Average has gone from +9 on 5/1 to +2.50 as of yesterday, to a tie today.

Currently I have Wisconsin as a swing state leaning Obama meaning right now I am giving him Wisconsin's electoral votes in our EV count. For now I am going to leave that alone, but one more poll within the MOE that reinforces this trend and I will officially move Wisconsin from "leans Obama" to "toss up". If that happens Obama's EV total will go from 212 secure / 247 with leaners to 212 secure / 237 with leaners and the toss up EV total will go from 52 to 62. Romney's numbers in that event will not change but will stay at 170 secure / 239 with leaners.

I am still waiting on polls for New Mexico and New Hampshire that may also result in a column shift. To reiterate both states only have one poll in the period we are calculating and both are PPP polls so until I get some data that supports or refutes those polls from multiple agencies they will be left alone at New Mexico "leans Obama" and New Hampshire "leans Romney".

No other EV shifts seem justified at this time. That leaves Colorado (9), Iowa (6), Nevada (6), Ohio (18), and Virginia (13) in the "toss up" category with 52 EVs at stake. "Leaning Obama" is New Mexico (5), Pennsylvania (20), and Wisconsin (10) for 35 EVs. "Leaning Romney" is Florida (29), Missouri (10), New Hampshire (4), and Arizona (11) at 69 EVs.

 

[Montrovant]

A respectful request to NYCarbineer, we are having a perfectly civil discussion here on the most current polling data available and what that probably means in relation to what we know of previous races and within the current dynamics.

If you have a problem with BluePantom, please take it to the Flame Zone or someplace else, but please don't mess up what the rest of us are enjoying here. BP has done a yeoman's job of compiling and posting the figures and he has adequately and competently sourced every one. You can contribute to the thread by showing compelling data with links that would challenge what he posts, or your personal sense of what the numbers mean, or your perception of polls in general, but school yard taunts are not appropriate here.

We have closed threads now here at USMB? And you're a moderator?

What part of her making 'a respectful request' leads to this comment? Foxfyre has asked that you try to refrain from making this into a pissing contest between you and BP. I agree with her request; I'd like this thread to remain not only civil but as objective as possible. You and BP clearly don't get along, so if you could try to avoid knocking heads with him (and he with you) it would be appreciated.

You've made it clear you think he is being partisan and dishonest. The rest of us reading the thread can see that, and can form our own opinions about his work here. What I think Foxfyre was saying, and what I'll reiterate, is that we are enjoying the thread and the work BP has put into it. There's nothing wrong with pointing out where you think his methods are wrong, but the manner in which you have gone about it (and, to be fair, the way his replies to you have been put) is not in the spirit of the thread.

 

[NYcarbineer]

A respectful request to NYCarbineer, we are having a perfectly civil discussion here on the most current polling data available and what that probably means in relation to what we know of previous races and within the current dynamics.

If you have a problem with BluePantom, please take it to the Flame Zone or someplace else, but please don't mess up what the rest of us are enjoying here. BP has done a yeoman's job of compiling and posting the figures and he has adequately and competently sourced every one. You can contribute to the thread by showing compelling data with links that would challenge what he posts, or your personal sense of what the numbers mean, or your perception of polls in general, but school yard taunts are not appropriate here.

We have closed threads now here at USMB? And you're a moderator?

What part of her making 'a respectful request' leads to this comment? Foxfyre has asked that you try to refrain from making this into a pissing contest between you and BP. I agree with her request; I'd like this thread to remain not only civil but as objective as possible. You and BP clearly don't get along, so if you could try to avoid knocking heads with him (and he with you) it would be appreciated.

You've made it clear you think he is being partisan and dishonest. The rest of us reading the thread can see that, and can form our own opinions about his work here. What I think Foxfyre was saying, and what I'll reiterate, is that we are enjoying the thread and the work BP has put into it. There's nothing wrong with pointing out where you think his methods are wrong, but the manner in which you have gone about it (and, to be fair, the way his replies to you have been put) is not in the spirit of the thread.

I pointed out that he lied to try to make a point. Most of you probably didn't even bother to check to see whether he lied or not; I happen to know he lied because I know the history of polling better than he does. You people are making a federal case of me pointing out that he LIED,

while treating him as some of genius/expert on polling.

This poster is not an 'analyst'. He is a rightwing partisan propagandist. The fact that he is dressing up his propaganda in a fancy costume does not change that fact.

 

[NYcarbineer]

The latest Foxnews poll has Obama up 46 to 39. That's from 46 to 46 3 weeks ago.

Romney lost his points to 'other' or 'undecided'. That makes sense. The more people are finding out about Romney the less appealing he is.

 

[BluePhantom]

So I will be adding a new breakdown report next week as well as the standard reports. This will be the Likely Voter Report. As we are seeing more and more LV polls and they are more accurate I will include a new category that considers just those polls. There will be three sub-categories: Verifiable, Estimated, and Combined. Verifiable is when the poll is officially released as LV on RCP. For Estimated I will take those RV polls that have enough demographics and crosstab information to plug into a generic 32/32/36 model and convert it to a generic LV poll. I will also be tracking the LV breakdown of D/R/I where possible over time to adjust this generic model according to current trends if I can get enough consistent data to do so. Then I will provide a Combined average of Verifiable and Estimated. I will do this in both Base Average (not adjusted for sample size) and Adjusted Average (adjusted for sample size) formats as well as a two week rolling average of both.

So basically I am doubling the size of the weekly report in order to distinguish between RV and LV polls as well as between media and professional polls. For now the charts on my spreadsheet will continue to reflect both RV and LV models combined. If I am getting enough data on a consistent basis to change them to a strict LV model I will do that in the future.

If you have suggestions, questions, or concerns now is the time to speak up as I am going to have to write a lot of sub-routines into the spreadsheet and once I have them done I aint changing them without a damned good reason.

 

[NYcarbineer]

I, on the other hand, will be enjoyably concise, and pleasantly unpretentious.

 

[nodoginnafight]

I, on the other hand, will be enjoyably concise, and pleasantly unpretentious.

Holy crap - I just spit coffee out of my nose.

Amazing how many words you have to use to perform statistical alchemy.

 

[NYcarbineer]

And meanwhile, President Obama moves out to a 3 point lead,

on Rasmussen. 11 point swing in a week if I'm not mistaken.

Analyze that!