ReadabilityThe elephant in the polling numbers
I’ve written a lot about the skew in the polls but today in the hot air green room Matt Vespa in the process of debunking Chris “The sky is falling for the GOP” Cilliza brings up a point that almost nobody is talking about:
This leads us to the next poll: the direction of the country. What “jolt of optimism” is Cillizza talking about? Rasmussen has 36% of Americans believing the nation is on the right track.
Now while we give our friends on the left a few minutes to scream how: “Rasmussen can’t be trusted” business lets take a look at the overall numbers in this poll since Jan 2009 the month Barack Obama Took office:
When you look at that chart remember this is an AVERAGE of every single polls that is measured not a single poll that skews one way or the other.
Right now the MSM is running a ton of polls with huge Democrat skews yet take a look at the results at the end of the chart for today:
56.3 vs 37.6 Wrong track over right track that’s a spread of 18.7 and consider this: The CBS/NYT poll and the NBC/WSJ poll, the two polls that are the most skewed have a wrong track/right track numbers of –19 & –16 respectively.
Think about that: Even with a sample that couldn’t favor Obama more if they were paid employees of the white house they can’t even get a gap of less that 10 pts let alone a favorable number.
More importantly consider the historical perspective:
You have to go to June 13th 2009 to find a day during the last 4 years where the right track/wrong track numbers were equal.
On Jan 20th 2010 the national spread was 56.9 to 36.6 just 2.4 off today’s number yet the day before Massachusetts, a state that nobody contends is a state where the president has a chance of losing in 2012 elected a Republican senator to replace the late Ted Kennedy.
On Election day 2010 the spread of 63.8 to 31.2 A 32.6 points gap, just under double what it is today and the GOP won nationwide in an election so historic it reached into statehouses all over the country.
On September 13th 2011 the spread was 73.6 to 20.8 a gap of 52.8 and for the first time since Woodrow Wilson was president a republican won an election in the 9th district of NY right in the heart of New York City.
What does that mean for today?
On Election day 2012, It is unlikely the right direction/wrong direction spread will be 52.8. That suggest Mitt Romney isn’t going to manage to win the State of New York.
On Election day 2012 It is unlikely the right/wrong spread will be 32.6. That suggests Mitt Romney isn’t going to manage a victory of historic proportions all over the country.
But with just over 40 days to the election with polls so skewed it’s a wonder the computers they are tabulated on don’t tip over the right/wrong track is 18.7 pretty close to the numbers on the Day Scott Brown was first elected, does that mean Mitt is going to do well enough to take a state as blue as Massachusetts?
I think not, but unfortunately for the left, he doesn’t have to take Massachusetts, or New York or any other deep blue state. He has to take the states in the middle, the ones that once voted for the GOP but went to Obama in the heady days of Nov 2008 for the left.
And a with a right track /wrong track number of 18.7 that’s not only doable that’s practically inevitable.
I’m sure there are people who might not like Mitt Romney, I’m sure there are those who don’t like he is a Mormon, those who don’t like he is rich and those who think he’s kind of stiff.
But unless you live in the land of deep blue delusion nobody in their right mind would suggest he is unqualified to be president.
This election isn’t going to be about: “Do I like Mitt Romney?” This election isn’t even going to be about “Has Barack Obama done a good job” the polling shows the verdict is already in on it.
The people are dying to vote against Barack Obama, all Mitt Romney has to do is convince them he can do the job, because Barack Obama has already convinced them he can’t.
Update: VDH notes how few votes it would have taken to make 1980 a Carter victory
In other words, until the very last week of the campaign, Reagan had an uphill fight. True, he eventually won a landslide victory in the Electoral College (489 to 49) and beat Carter handily in the popular vote. Yet Reagan only received a 51-percent majority.
What had saved Reagan from a perfect storm of negative factors — gaffes, additional conservative candidates on the ballot, a single debate, and a biased media — was not just the debate. Voter turnout was relatively low at only 53 percent. If Reagan’s conservative base was united and energized, Carter’s proved divided and indifferent.
Mitt doesn’t have a John Anderson but he also doesn’t have Reagan’s charisma, his conclusion:
The winner probably won’t be decided by old video clips, gaffes, or even campaign money, but by turnout and the October debates — depending on whether incumbent Obama comes across as a petulant Carter and challenger Romney appears an upbeat Reagan. As in 1980, voters want a better president — but they first have to be assured he’s on the ballot.
I’ve written a lot about the skew in the polls but today in the hot air green room Matt Vespa in the process of debunking Chris “The sky is falling for the GOP” Cilliza brings up a point that almost nobody is talking about:
This leads us to the next poll: the direction of the country. What “jolt of optimism” is Cillizza talking about? Rasmussen has 36% of Americans believing the nation is on the right track.
Now while we give our friends on the left a few minutes to scream how: “Rasmussen can’t be trusted” business lets take a look at the overall numbers in this poll since Jan 2009 the month Barack Obama Took office:
When you look at that chart remember this is an AVERAGE of every single polls that is measured not a single poll that skews one way or the other.
Right now the MSM is running a ton of polls with huge Democrat skews yet take a look at the results at the end of the chart for today:
56.3 vs 37.6 Wrong track over right track that’s a spread of 18.7 and consider this: The CBS/NYT poll and the NBC/WSJ poll, the two polls that are the most skewed have a wrong track/right track numbers of -19 & -16 respectively.
Think about that: Even with a sample that couldn’t favor Obama more if they were paid employees of the white house they can’t even get a gap of less that 10 pts let alone a favorable number.
More importantly consider the historical perspective:
You have to go to June 13th 2009 to find a day during the last 4 years where the right track/wrong track numbers were equal.
On Jan 20th 2010 the national spread was 56.9 to 36.6 just 2.4 off today’s number yet the day before Massachusetts, a state that nobody contends is a state where the president has a chance of losing in 2012 elected a Republican senator to replace the late Ted Kennedy.
On Election day 2010 the spread of 63.8 to 31.2 A 32.6 points gap, just under double what it is today and the GOP won nationwide in an election so historic it reached into statehouses all over the country.
On September 13th 2011 the spread was 73.6 to 20.8 a gap of 52.8 and for the first time since Woodrow Wilson was president a republican won an election in the 9th district of NY right in the heart of New York City.
What does that mean for today?
On Election day 2012, It is unlikely the right direction/ wrong direction spread will be 52.8. That suggest Mitt Romney isn’t going to manage to win the State of New York.
On Election day 2012 It is unlikely the right/wrong spread will be 32.6. That suggests Mitt Romney isn’t going to manage a victory of historic proportions all over the country.
But with just over 40 days to the election with polls so skewed it’s a wonder the computers they are tabulated on don’t tip over the right/wrong track is 18.7 pretty close to the numbers on the Day Scott Brown was first elected, does that mean Mitt is going to do well enough to take a state as blue as Massachusetts?
I think not, but unfortunately for the left, he doesn’t have to take Massachusetts, or New York or any other deep blue state. He has to take the states in the middle, the ones that once voted for the GOP but went to Obama in the heady days of Nov 2008 for the left.
And a with a right track / wrong track number of 18.7 that’s not only doable that’s practically inevitable.
I’m sure there are people who might not like Mitt Romney, I’m sure there are those who don’t like he is a Mormon, those who don’t like he is rich and those who think he’s kind of stiff.
But unless you live in the land of deep blue delusion nobody in their right mind would suggest he is unqualified to be president.
This election isn’t going to be about: “Do I like Mitt Romney?” This election isn’t even going to be about “Has Barack Obama done a good job” the polling shows the verdict is already in on it.
The people are dying to vote against Barack Obama, all Mitt Romney has to do is convince them he can do the job, because Barack Obama has already convinced them he can’t.
Update: VDH notes how few votes it would have taken to make 1980 a Carter victory
In other words, until the very last week of the campaign, Reagan had an uphill fight. True, he eventually won a landslide victory in the Electoral College (489 to 49) and beat Carter handily in the popular vote. Yet Reagan only received a 51-percent majority.
What had saved Reagan from a perfect storm of negative factors — gaffes, additional conservative candidates on the ballot, a single debate, and a biased media — was not just the debate. Voter turnout was relatively low at only 53 percent. If Reagan’s conservative base was united and energized, Carter’s proved divided and indifferent.
Mitt doesn’t have a John Anderson but he also doesn’t have Reagan’s charisma, his conclusion:
The winner probably won’t be decided by old video clips, gaffes, or even campaign money, but by turnout and the October debates — depending on whether incumbent Obama comes across as a petulant Carter and challenger Romney appears an upbeat Reagan. As in 1980, voters want a better president — but they first have to be assured he’s on the ballot.
Like this:
Like Loading...
Related
[…] Rather than looking at the fact that every single one of these polls show the public thinking the country is on the wrong track, rather than looking at the registration trends over the last 9 years as I did, rather than […]
[…] universe“, the same media that called us conspiracy theorists when we dared point out the mathematics problems with the polls, some even admitted by pollsters. The same media that pooh poohed the nine […]
I don’t care a rat’s patootie about Romney’s likeability (or the real or perceived likeability of any candidate in any contest). I’m voting for him because he’s not a Muslim and he’s not a babykiller.
Datechguy – I have read a lot about polling bias over many different sites. One thing I don’t see a lot of thought-provoking ideas about is HOW the samples are getting so heavily skewed to D+”Whatever” by all the supposedly biased organizations conducting them. While I don’t delude myself into thinking that most mainstream media outlets are unbiased, I have a hard time believing that they are INTENTIONALLY conducting polls in such a way to result in D-skewed responders and getting away with it. (I.e., How could the WSJ allow nefariously D-slanted poll under its sponsorship??) Is there a technical answer to this from the polling profession that makes it OK to skew the sample of responders to D? You seem to be much more knowledgable/adept than I on this subject — appreciate your thoughts on the ‘WHY’ on these skewed polls, not the usual ‘WHAT’ it might mean. Thx.
Zombie has a fair piece on that but the why’s could be several things.
For those not deliberately trying to create propaganda, it could simply be they are using the 2008 models and until there have a new election they will not change.
Alternatively for those not trying to affect the election as the Democrats had a registration advantage for, well just about forever until 2010 this bias might be specifically built into their computer models.
Now those trying to skew the election might also have these situations but refuse to alter their models to achieve a goal, and considering the amount of money involved in an election it is frankly odd to believe that people would not be deliberately trying to manipulate the results for their own ends.
However it’s a great question and I think I’ll turn it into a full post, Thanks for reading Doc.
“using the 2008 models”? You must be referring to a practice whereby pollsters don’t just take answers from n-number of responders and report the data. They have to “adjust” it so they think it fits the electorate at large? And they do this primarily what they think the actual split is of R v D v I? Do they not disclose what their “weighting factor” is when they have to make subjective adjustments like this? Crazy.
One thing that has not been discussed much in this debate is the percentage of people who actually respond to pollsters. I did see a figure of 9% somewhere. If true, that means that 9 out of 10 people contacted decline to participate, for whatever reason. Seems like there is a lot of self selection going on, which is the reason internet polling tends to have more entertainment value than statistical validity. The significant question would be are lefties more willing to participate in polls than righties? [my completely unscientific gut feel tells me that that may in fact be the case, given the amount of positive validation the left gets in the media and the culture in general, whereas conservatives are used to keeping their opinions more close-hold] If the answer is yes, that would explain a tendency to skew the samples that does not implicate pollsters in “cooking the books”, at least not directly. A study could be done pretty easily on this. Get names and phone numbers of a sufficient sample size of known registered Dems and Republicans and conduct a “survey”, the real object of which is only to determine if respondents will participate, and then do some analysis to determine if any differences are statistically significant. Maybe a study like that already exists out there somewhere.
I don’t know I always answer polls, I find them fascinating and I like to listen to the pollster when they ask a question to try to figure out their beliefs
[…] DaTechGuy analyzes RCP’s right track/wrong track numbers. Right now the MSM is running a ton of polls with huge Democrat skews yet take a look at the results at the end of the chart for today: […]
Thank you for clarity. But races are still nail biters (hoof biters as the case may be) for me.
Off topic but: http://evilbloggerlady.blogspot.com/2012/09/elizabeth-warrens-supporters-are.html You might want to look in and see what Professor Jacobson is posting about that Senate race in your state.
[…] Cross-Posted from my blog jQuery(document).load(function(){ stLight.options({publisher:'506523406'}); });emailprintvar dd_offset_from_content = 40; var dd_top_offset_from_content = 0;0 – I recommend This Share This Article Related News About Author datechguy […]