ReadabilityDemoralized as Hell, Morning Joe & the Last Gasp Poll before Realville
I haven’t dived deeply into polls much lately because frankly there has been no point, Mitt Romney has this race and baring the “Live Boy Dead girl” scenario it’s not changing. (actually considering Romany’s rep it wouldn’t take a dead girl, a live girl would shatter his clean image completely)
But with one week to go the last gasp effort to rally the MSNBC troops took place today on Morning Joe as they touted the Quinnipiac/CBS?NYT polls Saying Obama is up 5 in Ohio with a week to go.
What really got me was their claim that the Quinnipiac poll was the “most accurate” and again playing the “poll denier” card. I’ve personally found their numbers the least believable and while they were talking about the fights in other states as an alternative to Ohio for Romney I looked up the internals of this Ohio/VA/Florida poll and found this:

After talking about how reliable Quinnipiac is, for 15 minutes and while I was tweeting out the D+8 samples in all of these states D+7 in Florida, they suddenly pivoted as Mark Halperin brought up the D+8 sample and people asked how can this be the case if Mitt is up by huge margins among independents in this poll?
At this time they made the case that the split in the sample defines the electorate while I argue that the accurate of the poll is based on how the splits match the electorate.
Now Polls have limitation based on the response rate and the various methods of getting people to answer but the actual registration and demographics of a state are a reality, they may change over time but they are what they are no matter what but of a sample is used.
In the 7 AM Hour more of the same and in the 8 AM hour they repeated this nonsense. This isn’t a poll, this is a last gasp before Media Credibility day arrives and it becomes impossible to deny what already exists.
That the Morning Joe team is spending a show trying to sell the accuracy of a D+8 poll in Ohio & VA & +7 in Fla to their far left base speaks volumes, here is what it says to me:
When you are reduced to Chuck Todd arguing body language in private conversations that’s pretty bad, by contrast you have Romney people canvasing during the damn storm.
Folks, there is going to be a turnout effort in Ohio like you’ve never seen. George Cullen has been canvassing door-to-door for the past five weeks. He usually hits about 50 doors a day, but only got 25 under the storm conditions yesterday. He’s a West Point graduate who served six years in the Army, and he says, “The conservatives are very energized.”
Yeah this is a D+8 state SURE! It will be fun to watch these guys next Wednesday.
Update: Just called Robert Stacy McCain on the ground in Ohio and asked about the “Body Language” stuff from Chuck Todd: “It’s crazy” He spoke to Kevin Madden on Sunday one of the top guys in the Romney Campaign after rally. He seemed calm and laid back. He hasn’t seen any of this scary “Body Language” from the Romney folks.
They are still working hard, they have not slacked off nor would anyone expect them to but if our friends on the left want to tell themselves this, as the folks in the south say “Bless their Hearts”.
Update 2: Ed Morrissey elaborates:
In each of these three states, the CBS/NYT/Q-poll shows Republicans at a lower percentage level of turnout than in the 2008 election. If one makes that assumption, it’s not too difficult to be guess that Obama might be ahead. However, that’s exactly the opposite of what all other polls rating enthusiasm are telling us what the electorate will look like on Tuesday. In fact, it’s not even what this poll shows, with Republican enthusiasm +16 over Democrats in Florida, +14 in Ohio, and +7 in Virginia.
I guess that explains why we didn’t see Ed on the show today. I suspect that message wouldn’t have been all that welcome.
Update 3: Instalanche and Michael Graham who you might recall was the owner of the site “Anybody But Mitt” says what is coming:
I believe we’re on the verge of a solid Romney win for two reasons. One is the objective evidence. The other is the ugly desperation of the Obama campaign in its final days.
and he raises a very important point I never thought of:
The president, on the other hand, has peaked at 47 percent. The Battleground Poll model shows Obama losing 52 percent to 47 percent. Rasmussen daily tracking has Obama losing 49 percent to 47 percent. Pew has him tied: 47 percent to 47 percent. But more important, all the polls show Obama sliding or stuck. None show any upward movement.
Obama supporters are quick to tell you “the only poll that matters is the one on Election Day.” Two things: a) that’s what candidates who are behind always say; b) this is election day.
Next Wednesday is going to be fun.
Update 4: Stacy who I so rudely woke up in the first update asks an obvious question:
You might think some reporter at the New York Times, which co-sponsored this poll, would think it worth his while to interview the people at Quinnipiac and ask where they’re getting these weird over-samples of Democrats. What is causing this? How is it that Republicans report greater voting enthusiasm by such large margins, and yet are underrepresented in the poll’s sample? It makes no sense.
I’d love to hear the answer.
BTW PPP poll now being touted, Iowa Sample is D+5 Wisconsin D+4
I haven’t dived deeply into polls much lately because frankly there has been no point, Mitt Romney has this race and baring the “Live Boy Dead girl” scenario it’s not changing. (actually considering Romany’s rep it wouldn’t take a dead girl, a live girl would shatter his clean image completely)
But with one week to go the last gasp effort to rally the MSNBC troops took place today on Morning Joe as they touted the Quinnipiac/CBS?NYT polls Saying Obama is up 5 in Ohio with a week to go.
What really got me was their claim that the Quinnipiac poll was the “most accurate” and again playing the “poll denier” card. I’ve personally found their numbers the least believable and while they were talking about the fights in other states as an alternative to Ohio for Romney I looked up the internals of this Ohio/VA/Florida poll and found this:

After talking about how reliable Quinnipiac is, for 15 minutes and while I was tweeting out the D+8 samples in all of these states D+7 in Florida, they suddenly pivoted as Mark Halperin brought up the D+8 sample and people asked how can this be the case if Mitt is up by huge margins among independents in this poll?
At this time they made the case that the split in the sample defines the electorate while I argue that the accurate of the poll is based on how the splits match the electorate.
Now Polls have limitation based on the response rate and the various methods of getting people to answer but the actual registration and demographics of a state are a reality, they may change over time but they are what they are no matter what but of a sample is used.
In the 7 AM Hour more of the same and in the 8 AM hour they repeated this nonsense. This isn’t a poll, this is a last gasp before Media Credibility day arrives and it becomes impossible to deny what already exists.
That the Morning Joe team is spending a show trying to sell the accuracy of a D+8 poll in Ohio & VA & +7 in Fla to their far left base speaks volumes, here is what it says to me:
When you are reduced to Chuck Todd arguing body language in private conversations that’s pretty bad, by contrast you have Romney people canvasing during the damn storm.
Folks, there is going to be a turnout effort in Ohio like you’ve never seen. George Cullen has been canvassing door-to-door for the past five weeks. He usually hits about 50 doors a day, but only got 25 under the storm conditions yesterday. He’s a West Point graduate who served six years in the Army, and he says, “The conservatives are very energized.”
Yeah this is a D+8 state SURE! It will be fun to watch these guys next Wednesday.
Update: Just called Robert Stacy McCain on the ground in Ohio and asked about the “Body Language” stuff from Chuck Todd: “It’s crazy” He spoke to Kevin Madden on Sunday one of the top guys in the Romney Campaign after rally. He seemed calm and laid back. He hasn’t seen any of this scary “Body Language” from the Romney folks.
They are still working hard, they have not slacked off nor would anyone expect them to but if our friends on the left want to tell themselves this, as the folks in the south say “Bless their Hearts”.
Update 2: Ed Morrissey elaborates:
In each of these three states, the CBS/NYT/Q-poll shows Republicans at a lower percentage level of turnout than in the 2008 election. If one makes that assumption, it’s not too difficult to be guess that Obama might be ahead. However, that’s exactly the opposite of what all other polls rating enthusiasm are telling us what the electorate will look like on Tuesday. In fact, it’s not even what this poll shows, with Republican enthusiasm +16 over Democrats in Florida, +14 in Ohio, and +7 in Virginia.
I guess that explains why we didn’t see Ed on the show today. I suspect that message wouldn’t have been all that welcome.
Update 3: Instalanche and Michael Graham who you might recall was the owner of the site “Anybody But Mitt” says what is coming:
I believe we’re on the verge of a solid Romney win for two reasons. One is the objective evidence. The other is the ugly desperation of the Obama campaign in its final days.
and he raises a very important point I never thought of:
The president, on the other hand, has peaked at 47 percent. The Battleground Poll model shows Obama losing 52 percent to 47 percent. Rasmussen daily tracking has Obama losing 49 percent to 47 percent. Pew has him tied: 47 percent to 47 percent. But more important, all the polls show Obama sliding or stuck. None show any upward movement.
Obama supporters are quick to tell you “the only poll that matters is the one on Election Day.” Two things: a) that’s what candidates who are behind always say; b) this is election day.
Next Wednesday is going to be fun.
Update 4: Stacy who I so rudely woke up in the first update asks an obvious question:
You might think some reporter at the New York Times, which co-sponsored this poll, would think it worth his while to interview the people at Quinnipiac and ask where they’re getting these weird over-samples of Democrats. What is causing this? How is it that Republicans report greater voting enthusiasm by such large margins, and yet are underrepresented in the poll’s sample? It makes no sense.
I’d love to hear the answer.
BTW PPP poll now being touted, Iowa Sample is D+5 Wisconsin D+4
“BTW PPP poll now being touted, Iowa Sample is D+5 Wisconsin D+4″
Yes, and both polls claimed that Obama topped Romney among Independents by 7%. Right.
I hope that after this one is over, there’s a “Night of the Long Knives” for many a pollster.
Hi everyone. I see this conversation over and over: the conservatives claiming there is some conspiracy among the pollsters. There is not. The explanation for the polls with a big D# is very simple.
The pollsters do NOT adjust for voter ID. (Well, at least one does, but they tell you that openly.) We all know they ask questions to screen for likely voters. The D/R/I numbers that seem out of whack are due to differing likely voter screens. This year, the pollsters are really far apart with their likely voter models.
Some, like Mason-Dixon, Gallup and Rasmussen, use a stricter method to determine likely voters, and this method adjusts turnout to roughly the same as 2008, which obviously leads to a Romney win, since Romney’s picking up the independents. Other pollsters, like Ipsos, Quinnipiac and PPP, are using a model to determine likely voters that is looser, and it shows a group of D voters growing enough to offset the loss of the independents. That is the Hispanic vote.
So, simply put: if the Hispanic vote is the same size as previous elections, Romney wins. If it grows as some pollsters predict, Obama wins. Which side will be right? Well, if I could predict the future, I’d be writing this from my yacht while my trophy wife gave me a lap-dance. However, in 2008, who was the closest of all the pollsters: Scott Rasmussen. If I have to pick the best likely voter model, I unequivocally pick Scott Rasmussen.
One final point: if the Rs don’t get off their dainty derrieres and make Hispanics job #1, very soon Florida, Colorado and Arizona will be like Calif: unwinnable, and then we will have one-party rule.
re:polls
Wouldn’t a poll that showed my opponent up by 5 pts be used as motivation for the unlikely voter on the opposite side to vote? AND if my candidate was up 5 wouldn’t it cause my candidate’s voters to lax and not vote.
See: MA Special Election Brown vs. Coakley.
The Dem 7+ or 8+ oversample appears to be the average that the three legacy alphabet nets use to juice O’s numbers. Vide ABC & NBC along with their prestige sidekicks. Third time I’ve checked and weirdly—NOT!—the numbers are almost always in that range for all three on their last three forays into hoodwinkery.
I was also industrious enough to scroll down to page 18 [of 18] to check on party affiliations, and I found Virginia in this Q poll had Dem 35%, GOP 29%, Ind 31%. Not one of the news organizations in the MSM will bother to note this blatant attempt to hoodwink the casual reader, period. Given that info, Allen may be slightly ahead, and Romney is probably ahead.
Upchuck Todd lost it for a day after the first debate & appeared even-handed, was threatened by Brian the disciplinarian with a session with the leather strop, and is now reliably back on auto-zombie.
Romney will win Ohio. No way will it be D. I reckon 53/47.
[...] [...]
From the original post.
” How is it that Republicans report greater voting enthusiasm by such large margins, and yet are underrepresented in the poll’s sample?”
Why should greater enthusiasm necessarily translate into greater numbers? Republicans might look forward to voting against Obama with a burning intensity, whereas Dems will vote for him with more subdued emotions, but why do you think that necessarily means that there would be more Republicans than Democrats?
Sorry for the language, but the information is good news.
http://datechguyblog.com/2012/10/31/demoralized-as-hell-morning-joe-and-the-last-gasp-poll-before-realville/
@john pearson
Reputable pollsters do not use “turnout models”. They weight their raw results by certain relative constants – sex, race, class ratios etc. but not by Party ID. Party ID is almost as much a variable (to be polled for) as is candidate preference. The Party ID numbers they report is what they have found, not what they design.
It makes no sense for you to claim that Obama losing independents guarantees he will lose. The very same polls that tell you he is losing independents tell you that he is tied overall. They are giving you clear evidence that the one result does not determine the other.
The recent Pew poll that had the candidates tied, also had more Republicans than Democrats – how do you explain that?
I am not quite sure what the point of us arguing these matters is. The real world will intrude and settle these issues soon enough.
If some number of Republicans choose not to identify themselves as such, but claim to be independents, yet do support Romney, then you can easily have this phenomenon where a well-done poll shows both a smaller number of declared Republicans than expected, and a big lead for Romney amongst independents, and an overall tie.
Why would some number of Republicans claim to be independents? Perhaps these are moderate, squishy Republicans who feel quite uncomfortable with the hardline rightists who dominate the party, but who do feel that Romney is one of their own.
Seems more plausible than these nutty conspiracy theories….j
I suspect it’s a question of using a 2008 model
@ joe citizen:
Virtually every poll throughout this election cycle has used a 2008 Obama turnout model, in some cases, even higher Dem turnout than ’08. If Obama isn’t able to achieve those kinds of numbers, he WILL lose. The single greatest indicator of presidential election party ID breakdown is the previous midterm elections, These were dead-even in 2010.
According to virtually every poll, no matter how Obama-friendly, Obama will lose the Independent vote by between 8-20%. That alone is enough to sink him.
It also appears the Democrat base also faces a significant enthusiasm gap comparable to the GOP (between 5-15%). That, too, is probably enough to sink Obama.
According to both Gallup and Rasmussen, more voters in 2012 identify themselves as Republican than Democrat (Rasmussen has it approx. 2.6% more GOP, Gallup has it closer to a 1% advantage). These are enormous samples, with a far smaller margin of error than any IBD/TIPP poll Democrats are investing their hopes in. That, without the other 2 points, is enough to kill Obama’s chances.
Gallup’s survey of early voters (with a sample size of more than 3,000 people) has Romney up 52%-45%. That number sounds about right.
My question, Joe Citizen: will you see this as a mandate for Republican rule?
I live in Virginia. I can tell you that a D+8 in the Commonwealth is absolute insanity. In 2008, Obama’s margin was just over 6%. He is now thoroughly discredited with large groups of voters: whites, working class, small towns, suburbs, rural areas, veterans & military, business owners, independents. The Dems suffered losses in the gubernatorial and legislative races in 2009, in the congressional races in 2010, and again in legislative races in 2011. GOP has a registration advantage in the state over Dems. There’s no way they get even close to an 8+ party ID in Virginia.
Indeed, this IS election day and that was the Dem’s idea as well. I never favored early or gratuitous absentee voting. After all, it was invented for the military and they seem the one group targeted for suppression! But it seems it also is the playground of the fraudsters IF they don’t get out-performed even here. Romney is up 7-odd points in early voting exit polls, very close to his current high-end lead. 15% of projected votes have been cast. 15%! They will call a state with those numbers even for a Republican! And the Dems focus on early/abs. They were expected to LEAD here. Even the positive recent news may be understating things and here’s another bit of poll-dancing I have been unable to verify, perhaps Peter knows. Erick Erickson reported on his show some time ago that Obamaphones are explicitly opted IN to public phone lists, meaning pollsters. As most of you know, a cell phone cannot be legally called by unsolicited outfits like pollsters UNLESS you opt in. The O-phones are opted in and must be opted OUT. Given the huge explosion of O-phones (from 6k to 300k in Creverand, as we have been forced to learn) well, could the practical skew be more like a Dem +12? +14 as the most skewed poll I had it? And then there is the natural tendency of grumpy Righties to refuse to answer or even answer fraudulently, as I did the one time I was polled. That breaking down may be all the explanation necessary for the tightening polls and vastly UNDER predict the results next Tuesday. NEXT TUESDAY, DUDES! And you ladies too, of course.
For your final question, one explanation I’ve heard from the left is that Republicans are ashamed to call themselves Republicans and have started to refer to themselves as Independents. It neatly ties up both loose ends in one nice little package. Of course, there’s no evidence of this and it doesn’t square at all with 2010 or 2012 in Wisconsin, but they’re clinging to it.
[...] KoestlerFROM OHIO: POLLS, POLLS, POLLSPosted on | October 31, 2012 | No CommentsCINCINNATI, Ohio Pete Da Tech guy disassembles a Quinnipiac Poll that has samples of D+7 in Florida and D+8 in both F…, so that while Mitt Romney handily wins independent voters in both states, he nonetheless trails [...]
“b) this is election day.” Indeed. I just voted in MD.
Don’t count your chickens, but we could be in for a wave.
I live in Ohio. All of the all of the elected State officials are Republicans. The General Assembly and the Senate are Republican. Other than 2006-2010, it has been that way for the past 25 years. There is no way that Ohio is D+ anything. No Way.
[...] Peter Ingemi: Ride Right Through Them, They’re Demoralized As Hell! [...]
I was in OH over the weekend. We rented a bus to get there and drive us back and forth. That bus was the talk of the town. We stopped at a Walmart on our way home so we could buy storm supplies and people were coming up and talking to us. There is excitement in OH.