Green Room

National polls vs. Ohio polls: They can’t both be right

posted at 4:30 pm on October 28, 2012 by

Josh Jordan gauges the impending collision between the irresistible force of Romney’s national momentum and the immovable object of O’s small but stubborn lead in the Ohio polls. His take? The “immovable” object will move:

In Ohio, Republicans tend to outperform their share of the national vote: In the last six presidential elections, only in 2004 has the Republican candidate performed worse in Ohio than they did nationally, and even that was only a .3 percent difference. In the other five elections, the GOP candidate outperformed their margin of the national vote by an average of 3.1 percent. While it’s clearly possible for Republicans to perform worse in Ohio than nationally, it is very difficult to imagine a scenario where there is more than a point difference between them based on past elections. History would suggest Romney could not be up 2 points nationally while being down 2.1 in Ohio, which would mean Republicans would be under-performing in Ohio by over 4 points.

Democrats national turnout advantage is usually bigger than their Ohio turnout: Not since 1996, during Clinton’s reelection campaign, have Democrats had a larger turnout advantage in Ohio than nationally. In 2000 and 2008 Democrats were 2 points under their national turnout advantage and were actually 5 points under in 2004. Currently polls are showing an average Democratic turnout advantage of 6.3, which is 1.9 points higher than their current advantage nationally.

So why do most Ohio polls show O ahead? Because, says Jordan, there’s some subset of Democrats who are fibbing when asked by pollsters whether they’ve already voted. They know they’re supposed to — Obama’s been droning on about it nonstop — yet they haven’t. But when they lie and say they have, they automatically end up being counted as “likely voters,” which explains why you’re seeing Democrats oversampled in a lot of Ohio polls.

Really, this is common sense. How plausible is it that a guy who’s now below 47 percent in RCP’s national average is going to win a state that famously tracks with national sentiment each cycle? He’s getting hammered nationally with independents, especially on the core issue of the economy, and yet somehow he’s two points ahead overall in the ultimate bellwether? Really?

Recently in the Green Room:

Blowback

Note from Hot Air management: This section is for comments from Hot Air's community of registered readers. Please don't assume that Hot Air management agrees with or otherwise endorses any particular comment just because we let it stand. A reminder: Anyone who fails to comply with our terms of use may lose their posting privilege.

Trackbacks/Pings

Trackback URL

Comments

Anybody elses facebook share, window, pop up behind the text, making it impossible to share?

2Tru2Tru on October 28, 2012 at 4:35 PM

Only in Green Room, I should of added.

2Tru2Tru on October 28, 2012 at 4:37 PM

Kinda lonely in here.

2Tru2Tru on October 28, 2012 at 4:38 PM

The solo thread. Feels like a White democrat in south Carolina…. No?

Bensonofben on October 28, 2012 at 4:51 PM

Kinda like talking to yourself, except with typing!

kringeesmom on October 28, 2012 at 4:53 PM

So why do most Ohio polls show O ahead? Because, says Jordan, there’s some subset of Democrats who are fibbing when asked by pollsters whether they’ve already voted.

I think there are some who are fibbing an saying they are indys too. Might be why the indys suddenly shifted toward Obama in a recent poll. I haven;t had time to check to see if any nefarious forces are encouraging Democrats to lie to pollsters.

forest on October 28, 2012 at 4:56 PM

So I’m wondering aloud, what good does it do for democrats to fool themselves like this? Watching Ryan rally in Ohio, talking founding principals and stuff. Why won’t he address the Binder, Big Bird, Bayonet issues? What is he trying to hide?

Night Owl on October 28, 2012 at 4:59 PM

Fool me once — shame on you.

Fool me twice — shame on ME.

 

That is what it essentially and succinctly boils down to for each and every single legitimate voter in the entire United States of America.

FlatFoot on October 28, 2012 at 5:00 PM

Now here’s Romney. He’ll probably talk about jobs and stuff. What I want to know, is where are his tax returns?

Night Owl on October 28, 2012 at 5:03 PM

Allah, you sound positively optimistic. Careful, your reputation might get tarnished. ;)

tdpwells on October 28, 2012 at 5:04 PM

Is that optimism I hear creeping into the voice of that lovable Eeyore AP?

Nethicus on October 28, 2012 at 5:05 PM

Now here’s Romney. He’ll probably talk about jobs and stuff. What I want to know, is where are his tax returns?

Night Owl on October 28, 2012 at 5:03 PM

Locked up in the same vault as Obama’s college transcripts and financial aid information showing that he received grants/funding as a foreign student because he lied about being born in Kenya to get a leg up in the admissions process.

Or something.

tdpwells on October 28, 2012 at 5:07 PM

I think there are some who are fibbing an saying they are indys too. Might be why the indys suddenly shifted toward Obama in a recent poll. I haven;t had time to check to see if any nefarious forces are encouraging Democrats to lie to pollsters.

forest on October 28, 2012 at 4:56 PM

That poll was a CNN poll that had Obama winning indies in their last poll as well.

jarodea on October 28, 2012 at 5:11 PM

yes, and posting a Romney lead in ohio will in essence proclaim obama loss of the national election; no pollster wants to do that, they know they will get the wrath of axelturf if the do

runner on October 28, 2012 at 5:16 PM

tdpwells on October 28, 2012 at 5:07 PM

LOL!

Night Owl on October 28, 2012 at 5:18 PM

Really, this is common sense. How plausible is it that a guy who’s now below 47 percent in RCP’s national average is going to win a state that famously tracks with national sentiment each cycle? He’s getting hammered nationally with independents, especially on the core issue of the economy, and yet somehow he’s two points ahead overall in the ultimate bellwether? Really?

For those that cared, I took the liberty of seeing the internals of what I would consider the three “outliers” of the average: Time (Obama +5), CNN/OCP (Obama +4) and WCMH NBC4/SurveyUSA (Obama +3).

Time had a D37/R28/I29/O6 model. Can’t tell if it was all landline or not, but I would be surprised if it wasn’t the case.

CNN/OCP had a D35/R32/I33 model, all interviewed by landline.

WCMH NBC4/SurveyUSA had a D39/R32/I29 model. They also contacted people through landlines and through a pop-up questionnaire on smartphones or other mobile devices.

And for what it’s worth, the survey sample of the Ohio Poll (Obama-Romney tie) had over 1,015 respondents in a D47/R45/I10 model, conducted with landlines AND smartphones.

If you want my gut feeling, I’m siding with the Ohio Poll and Rasmussen who both have it as a virtual tie.

Myron Falwell on October 28, 2012 at 5:22 PM

Early voting definitely seems to be skewing some of these polls. I also think their likely voter screens are too loose in general, making them more akin to registered voter polls than anything else.

changer1701 on October 28, 2012 at 5:25 PM

This is why I think there will be real violence after Romney wins. Let’s flip the situation. Let’s say that all the polls had Romney up in Ohio by 2 or 3 points but the breakdown was R+5. We’d all be saying why are they trying to get our hopes up? But democrats aren’t all that intellectually honest, they just see a poll showing Obama up and they believe it. They will actually believe the election was stolen. Remember the humorous “Diebold stole the election” back in 2004? It will be twice that bad this time around.

Capitalist Infidel on October 28, 2012 at 5:29 PM

yes, and posting a Romney lead in ohio will in essence proclaim obama loss of the national election; no pollster wants to do that, they know they will get the wrath of axelturf if the do

runner on October 28, 2012 at 5:16 PM

Winner winner, chicken dinner.

bgibbs1000 on October 28, 2012 at 5:33 PM

For those that cared, I took the liberty of seeing the internals of what I would consider the three “outliers” of the average: Time (Obama +5), CNN/OCP (Obama +4) and WCMH NBC4/SurveyUSA (Obama +3).

Time had a D37/R28/I29/O6 model. Can’t tell if it was all landline or not, but I would be surprised if it wasn’t the case.

CNN/OCP had a D35/R32/I33 model, all interviewed by landline.

WCMH NBC4/SurveyUSA had a D39/R32/I29 model. They also contacted people through landlines and through a pop-up questionnaire on smartphones or other mobile devices.

And for what it’s worth, the survey sample of the Ohio Poll (Obama-Romney tie) had over 1,015 respondents in a D47/R45/I10 model, conducted with landlines AND smartphones.

If you want my gut feeling, I’m siding with the Ohio Poll and Rasmussen who both have it as a virtual tie.

Myron Falwell on October 28, 2012 at 5:22 PM

That Time poll is just flat out humorous! D+9 when the 2010 midterms were, I believe, R+1 in Ohio

The CNN poll looks ok as far as breakdown but they actually have Obama winning Independents by 6% when all other polls have Romney taking Indies by an average of over 10% so there is obviously something wrong with that one.

Once again the Survey USA poll has it D+7 so once again garbage in/garbage out.

Even the Ohio poll undersampled Indies by a bunch and if Romney is leading them by over 10% he would be leading that poll by 2 or 3 points.

Capitalist Infidel on October 28, 2012 at 5:53 PM

in one sense the leftist tilt of the polling helps Rs in the sense that it may cause marginal Ds not to vote because Ds will win anyway

the bad thing will be the hysterical reaction among the leftists in the press and elsewhere that the election has been ‘stolen’.

the press has revealed a staggering level of stonewalling the bengazhi stuff…so that they have so much skin in the left-wing camp, that they will do/say anything to de-legitimize conservatives.

they did this for 8 years against bush…but if mitt wins it will be worse. How to revive a free press will be a major project over the next decade

r keller on October 28, 2012 at 6:17 PM

I predict there will be a subset of fibbing libs that never make it to the polls on Nov 6. Barry’s just not worth wasting the gas on.

BettyRuth on October 28, 2012 at 6:25 PM

National polls vs. Ohio polls: They can’t both be right

Sure they can. Assume that Romney wins southern states like Texas and Georgia by lopsided (e.g. 78-22) margins, while losing traditionally sold blue states more narrowly (e.g. 53-47), and losing the swing states, including Ohio by margins like we see in the polling of 51-49. Tada.

Stoic Patriot on October 28, 2012 at 7:08 PM

And for what it’s worth, the survey sample of the Ohio Poll (Obama-Romney tie) had over 1,015 respondents in a D47/R45/I10 model, conducted with landlines AND smartphones.

I am no math major but something does not add up with their survey sample in that it adds up to 102

smokin hot politics on October 28, 2012 at 7:31 PM

While Rasmussen and Gallup have a reputation to keep up, these smaller state polling firms can easily massage the samples to get the result the want, by using unlikely voter turnout models or screening a bad registered voter batch.

Ohio will not be far off the national figures. 40 years of history tells the story. Case closed.

Norwegian on October 28, 2012 at 7:55 PM

The polls were scewed 2 years ago. Not the same. If gallup says Romney is far ahead in Ohio, they will get an injuction and sue Gallup. Gallup will let the voters be the final poll.

seven on October 28, 2012 at 7:56 PM

The pollsters are also making some odd mathematical assumptions about Ohio. They keep looking at the portion of their sample that has already voted, and have triangulated to determine how many early votes have been cast. That’d be a perfectly valid exercise, if most states didn’t already keep track of how many early votes have been cast. CNN for example, actually doubled the amount of early votes cast, which GREATLY threw off their numbers.

When you HAVE actual votes to compare your numbers to, you should check your numbers and weight them accordingly. Most aren’t, and its throwing off the polls considerably.

WolvenOne on October 28, 2012 at 8:21 PM

Romney is going to pick up over 300 electors on election night. Which is a stinging indictment of the other states which are populated by people who have no freakin idea how things work but like free stuff from the government.

Nethicus on October 28, 2012 at 8:22 PM

yes, and posting a Romney lead in ohio will in essence proclaim obama loss of the national election; no pollster wants to do that, they know they will get the wrath of axelturf if the do

runner on October 28, 2012 at 5:16 PM

^^^ This ^^^

However, the first one to do it will start the avalanche flowing.

mpthompson on October 28, 2012 at 8:23 PM

While Rasmussen and Gallup have a reputation to keep up, these smaller state polling firms can easily massage the samples to get the result the want, by using unlikely voter turnout models or screening a bad registered voter batch.

Ohio will not be far off the national figures. 40 years of history tells the story. Case closed.

Norwegian on October 28, 2012 at 7:55 PM

I remember in 2008 there were a lot of posters at hotair who were convinced that the polls were wrong and McCain was going to win. Quite honestly I’m perplexed, if Obama is down three points he shouldn’t be up in Ohio, but here we are. If Obama wins election his gotv effort is going to be legendary.

Ric on October 28, 2012 at 8:26 PM

That’d be a perfectly valid exercise, if most states didn’t already keep track of how many early votes have been cast. CNN for example, actually doubled the amount of early votes cast, which GREATLY threw off their numbers.

WolvenOne on October 28, 2012 at 8:21 PM

CNN doesn’t want to hear that people lie in polls if it doesn’t fit their narrative. Did you see the videos from Jimmy Kimmel where people were giving opinion on a debate that hadn’t happened yet? How many people lie to pollsters about whether they voted? Obviously a lot.

mpthompson on October 28, 2012 at 8:27 PM

Many of the Union members are lyin… They will tell their union bosses and most anyone else who ask yeah, Obama, but when they are in the booth alone many will be voting against the Coal Hater in Chief.

It’s their lives they are trying to save here.. Food on their table and shoes for their kids.

RockyJ. on October 28, 2012 at 8:37 PM

The Ohio variable is Obama’s GOTV on colleges.

The “already voted” polling distortion isn’t simple lying. A lot of these voters Barry is relying on think registering was their vote.

Honestly.

That’s why they herd people onto buses, like cattle. College students don’t know F*k aught about voting.

So you get a pollster who asks have you voted, and they think back to registration, assuming it was their vote.

Romney wins, we’ll hear the anecdotes and those will spun into voter suppression plights.

budfox on October 28, 2012 at 9:56 PM

Everybody seen this?

http://www.politico.com/news/stories/1012/82948.html

The closing paragraph:

I have always been a believer in data telling me the full story. Truth is, nobody knows what will happen on Election Day. But here is what we do know: 220,000 fewer Democrats have voted early in Ohio compared with 2008. And 30,000 more Republicans have cast their ballots compared with four years ago. That is a 250,000-vote net increase for a state Obama won by 260,000 votes in 2008.

Owen Glendower on October 28, 2012 at 9:56 PM

I’m happy we get to hear more from AP now that they’ve changed the Greenroom around.

Also, he has a different tone in here. I guess he gets to relax a little in the Greenroom. Maybe he’ll come in and talk to us once in a while.

The Hot Air Greenroom could be like the conservative twitter, but more exclusive. Exclusive of trolls and liberals in general. Hey! That’s a great idea! Part of the terms of use could be that if any of the liberals comments in teh Greenroom, he/she/it is banned!

And GET RID of those facebook comments in headlines.

jaime on October 28, 2012 at 11:50 PM

I wonder if national polling takes into account the massive wins Obama will have in California and New York. Every vote over victory is wasted. Does this skew the overall poll?

Dennis D on October 29, 2012 at 7:20 AM

The one plausible argument I’ve heard for why Obama could do better in Ohio than in the national:

Obama spent a fortune bombarding Ohio with negative ads during the period when Romney couldn’t respond (he was out of funds raised for the primary and couldn’t spend his main fund until after the convention).

Romney is now matching his spending, but Ohioans are getting sick and tired of the campaign and are tuning out the tv ads. In addition, it’s hard to shake a first impression, and Obama did manage to persuade a few percent of Ohioans that Romney would be even worse.

If that’s right, then we could see Romney win the national vote by several percent and still lose Ohio — but if Romney’s really up five points nationally (as I think he is) then he doesn’t need Ohio. Obama’s massive spending advantage in the pre-convention phase of the game didn’t extend to states like Wisconsin, Michigan, and Minnesota…

My head tells me Romney’s going to win big, but still I worry.

ClintACK on October 29, 2012 at 7:26 AM