If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Huffington Post)   After producing months of polls favoring Obama, Gallup decides to ask people who are actually voting in the election, and ... uh, well, umm ... yeah   (huffingtonpost.com) divider line 86
    More: Fail, Gallup, obama, audio file format, Pew Research Center, Sasha Issenberg, Ipsos, pollsters, combine  
•       •       •

6614 clicks; posted to Politics » on 09 Oct 2012 at 2:19 PM (1 year ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



86 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread

First | « | 1 | 2 | » | Last | Show all
 
2012-10-09 12:52:28 PM
In which "people who are actually voting in the election" actually means "people Gallup considers likely to vote in the election," which in turn means "a changing demographic that we can alter at will," which in turn means, "a target group malleable enough from sample to sample that we can consistently show results suggesting a narrow contest," which in turn means, "we stay marketable and our polling remains in high demand."
 
2012-10-09 01:20:59 PM
Done in one.
 
2012-10-09 01:21:27 PM

Pocket Ninja: In which "people who are actually voting in the election" actually means "people Gallup considers likely to vote in the election," which in turn means "a changing demographic that we can alter at will," which in turn means, "a target group malleable enough from sample to sample that we can consistently show results suggesting a narrow contest," which in turn means, "we stay marketable and our polling remains in high demand."


Ooo. Tho thynical.
 
2012-10-09 01:22:16 PM

Pocket Ninja: In which "people who are actually voting in the election" actually means "people Gallup considers likely to vote in the election," which in turn means "a changing demographic that we can alter at will," which in turn means, "a target group malleable enough from sample to sample that we can consistently show results suggesting a narrow contest," which in turn means, "we stay marketable and our polling remains in high demand."


That's some damn impressive skew. A Madison Avenue adjustment, if you will.
 
2012-10-09 01:32:19 PM
Nate Silver has been pointing out for a long time that Obama's lead was slimmer amongst likely voters vs registered voters, and many of the polls out there that he uses for his model have been using likely voters. The problem however, is its harder to figure out who likely voters are in July and August.

Gallup routinely uses registered voters as the metric until a month out, since its really difficult to predict likely voters so far out.
 
2012-10-09 01:40:35 PM
Does the sample include people at work or people with caller id?
 
2012-10-09 01:48:15 PM
The percentage of Americans likely to vote for whoever is irrelevant and I have no idea why they continue to repeat this pointless statistic.

/Electoral College.. how does it work?
 
2012-10-09 02:12:40 PM
And 8 yrs ago today: CNN poll shows Kerry inching past Bush, 49% to 48%, among likely voters.
 
2012-10-09 02:22:43 PM

neritz: And 8 yrs ago today: CNN poll shows Kerry inching past Bush, 49% to 48%, among likely voters.


Right. That means the polls are always wrong. Unless they say what we want them to say, and then they're right.

Also, Gallup puts Obama's approval rating today @ 53%, his second highest ever (the highest was last week @54%) . Dude's picking a pretty good time to be peaking.
 
2012-10-09 02:23:54 PM
oh sure libs first it's the unemployment rate isn't being made up, now when Obama wins, it's going to be nope there isn't wide spread voter fraud despite what the polls said that Romney is going to win.
 
2012-10-09 02:25:11 PM
The Mittmentum of Mittmania is pure Romdemonium!
 
2012-10-09 02:25:30 PM

Tymast: oh sure libs first it's the unemployment rate isn't being made up, now when Obama wins, it's going to be nope there isn't wide spread voter fraud despite what the polls said that Romney is going to win.


I'll think of a reply to this as soon as I figure out what you're trying to say.
 
2012-10-09 02:27:37 PM

UnholyMudcrab: Tymast: oh sure libs first it's the unemployment rate isn't being made up, now when Obama wins, it's going to be nope there isn't wide spread voter fraud despite what the polls said that Romney is going to win.

I'll think of a reply to this as soon as I figure out what you're trying to say.


:) ah just getting ahead of the curve , really isn't much to say to that.
 
2012-10-09 02:29:30 PM

Pocket Ninja: In which "people who are actually voting in the election" actually means "people Gallup considers likely to vote in the election," which in turn means "a changing demographic that we can alter at will," which in turn means, "a target group malleable enough from sample to sample that we can consistently show results suggesting a narrow contest," which in turn means, "we stay marketable and our polling remains in high demand."


Wham, bam, thank you ma'am, we are DONE here.

I have voted in every single election except one or two school board only ones since I was 18. Primaries, off Presidential years, everything.

I have never gotten polled. Ever. Although when I was in college my Mom did answer "likely Obama" on my behalf when the house was called, but that doesn't really count.

/behold, the power of only having a cell phone and being in my 20s!
 
2012-10-09 02:30:27 PM
Gallup's LV screen tends to be somewhat GOP leaning. Thus said, here is where we stand today:

Would Obama have lost if the election was held over this weekend, the time when the polls rolling in today were conducted? Most likely.
Would Obama lose if the election were today? Probably not, but it would be very close.
Will Obama lose in November? No, unless he farks up both of the next debates.
 
2012-10-09 02:30:41 PM
IT'S NOT FARK, IT'S TROLLS.COM
 
2012-10-09 02:30:45 PM

Rapmaster2000: The Mittmentum of Mittmania is pure Romdemonium!


Goddamn if this doesn't make me want to give advertisers money! Yee--hah!!!
 
2012-10-09 02:32:27 PM

Tell Me How My Blog Tastes: neritz: And 8 yrs ago today: CNN poll shows Kerry inching past Bush, 49% to 48%, among likely voters.

Right. That means the polls are always wrong. Unless they say what we want them to say, and then they're right.

Also, Gallup puts Obama's approval rating today @ 53%, his second highest ever (the highest was last week @54%) . Dude's picking a pretty good time to be peaking.


That definitely suggests that Romney will have a hard road ahead, since Romney is essentially running on an "anybody but obama" platform.
 
2012-10-09 02:34:45 PM

gilgigamesh: Tell Me How My Blog Tastes: neritz: And 8 yrs ago today: CNN poll shows Kerry inching past Bush, 49% to 48%, among likely voters.

Right. That means the polls are always wrong. Unless they say what we want them to say, and then they're right.

Also, Gallup puts Obama's approval rating today @ 53%, his second highest ever (the highest was last week @54%) . Dude's picking a pretty good time to be peaking.

That definitely suggests that Romney will have a hard road ahead, since Romney is essentially running on an "anybody but obama" platform.


Right now he's running on a "did you see how well I did in the debate?" platform. Effective for a few days, but that's it. If you look at the daily breakdown of the polls, it seems clear that he got a very tall but narrow bounce from the debate.
 
2012-10-09 02:36:39 PM

Tymast: UnholyMudcrab: Tymast: oh sure libs first it's the unemployment rate isn't being made up, now when Obama wins, it's going to be nope there isn't wide spread voter fraud despite what the polls said that Romney is going to win.

I'll think of a reply to this as soon as I figure out what you're trying to say.

:) ah just getting ahead of the curve , really isn't much to say to that.


I'll be honest. You had me going until I read it a second time to try and figure out why it was retarded, and realized I couldn't. Well done. That's high quality hurr.
 
2012-10-09 02:36:49 PM

ShawnDoc: Gallup routinely uses registered voters as the metric until a month out, since its really difficult to predict likely voters so far out.


That's *understandable*, I guess, but couldn't you just poll based on publicly available data of people who have managed to show up to the polls for 0/3, 1/3, 2/3, or 3/3 elections (generally breaks down to: rarely votes, just presidential years; president + house years; every random election there is, you be there)?

I mean, that's kind of how we prioritized our voter outreach, focusing on the most likely to show up (skipping over 'strongly identifies as [other party]' folks, of course) then moving down the list.

Don't forget to adjust for the young people who have only had very few (or none) elections *to* show up for, average the numbers over the past decade at most and not over the entire lifetime of the voter, and... yeah. Or are they (I assume) calling "likely voters" by those most likely to be energized enough to go to the polls for that particular election, cause eh, I'd be curious how that self-identification works as an accurate metric in the end.

/assume Silver has covered this
//but I'm over my BS NYT quota
///if you search an article from them via Google News you can get around it, but it's stupid, so I'm done bothering
 
2012-10-09 02:37:34 PM
So wait, Gallup apparently does this biannually; this is not just something they pulled out of their asses for this election.

My question is how this has caused the results to change in previous elections, and if they've made this change on similar dates, IE after the first debate. If they were suddenly making this shift to capitalize on the Romney blowout when they've made it earlier or later in previous years, then Pocket Ninja might actually have a point.

I don't really buy the "media wants to make it seem like a close race" meme, but I may be ignorant on history; God knows the media loves to hype up a story. Has this been the norm for the previous elections, or is it new to this one?
 
2012-10-09 02:38:46 PM

oldernell: Does the sample include people at work or people with caller id?


I suspect Gallup has less of a problem with that than some of the others. Gallup has a long and fairly respected history and I think people are more likely to pick up if they see Gallup on the caller ID than if they see a call coming from a Fox/CNN/NBC/NPR/Rasmussen/Jerry's window glass and polling.
 
2012-10-09 02:38:48 PM
I guess it's fortunate that the only votes that count in this election are the ones cast by the electoral college. And according to the link it's Obama 281 vs Romney at 191 with the winning number being 270. So, yeah, worried? Not even remotely.
 
2012-10-09 02:39:35 PM

StreetlightInTheGhetto: Pocket Ninja: In which "people who are actually voting in the election" actually means "people Gallup considers likely to vote in the election," which in turn means "a changing demographic that we can alter at will," which in turn means, "a target group malleable enough from sample to sample that we can consistently show results suggesting a narrow contest," which in turn means, "we stay marketable and our polling remains in high demand."

Wham, bam, thank you ma'am, we are DONE here.

I have voted in every single election except one or two school board only ones since I was 18. Primaries, off Presidential years, everything.

I have never gotten polled. Ever. Although when I was in college my Mom did answer "likely Obama" on my behalf when the house was called, but that doesn't really count.

/behold, the power of only having a cell phone and being in my 20s!


/behold being in the demographic least likely to vote
 
2012-10-09 02:39:54 PM
FWIW, four years ago, Gallup's switch to LVs increased McCain's numbers by six points.
 
2012-10-09 02:41:11 PM

spman: /behold being in the demographic least likely to voteexpensive to poll.


FTFY
 
2012-10-09 02:41:38 PM

oldernell: Does the sample include people at work or people with caller id?


Touche'

"Among likely voters who still have land lines and also call lawsuit settlement toll-free numbers..."
 
2012-10-09 02:42:57 PM

Mexican Jew Lizard: So wait, Gallup apparently does this biannually; this is not just something they pulled out of their asses for this election.

My question is how this has caused the results to change in previous elections, and if they've made this change on similar dates, IE after the first debate. If they were suddenly making this shift to capitalize on the Romney blowout when they've made it earlier or later in previous years, then Pocket Ninja might actually have a point.

I don't really buy the "media wants to make it seem like a close race" meme, but I may be ignorant on history; God knows the media loves to hype up a story. Has this been the norm for the previous elections, or is it new to this one?


I really wish 538 the actual website existed still.

Oh, hey, wayback machine:

here you go. If you had the time you could compare the individual poll data broken down by state (midway down page, right side) vs the actual result of the election and you could chart it by pollster and state over time leading up to the election.

I would because I generally geek out over this sort of thing, but that's about all the break I have from work now.

Damn, I miss 538 proper. The voter data guru in my state who we ended up buying really crappy data from (not entirely his fault because my district hadn't been contestable in the slightest until 2 years before I was there - not for decades - but it was misrepresented nonetheless) actually posted on there too. Never did get the nerve to call him out on that one, though. And his data was pretty damn reliable in other districts.

/shrug
 
2012-10-09 02:46:11 PM

Rapmaster2000: The Mittmentum of Mittmania is pure Romdemonium!


Romdemonium. That's good. I like that.
 
2012-10-09 02:46:58 PM
The only poll that counts is the popular vote, and even that can be circumvented by a governor who's your brother and a SCOTUS that was put in place by your father and his friends.
 
2012-10-09 02:47:00 PM

spman: /behold being in the demographic least likely to vote


You *do* know that you showing up to an election or voting by an absentee ballot is information that anyone can obtain, right? No one knows how you voted unless you answer a poll, but how historically likely you were to show up to vote is (relatively speaking) public knowledge.

I've seen the data that the Dems in my state have on me. Generally you want to encourage folks who have solid voting histories (I do) to vote... of course, I'm registered independent (so I fall off get out the vote rolls if I'm not polled). And of course, I don't have a land line.

Oddly enough. even the fact that I worked as a campaign coordinator for a freaking Democrat didn't show up in the system. I could've added that myself, but I didn't feel like doing work for them. So I was still listed as "maybe votes Democrat" last I checked, just from that single phone call my Mom answered in 2008.
 
2012-10-09 02:47:11 PM
Kodos.
 
2012-10-09 02:48:14 PM
Every day it becomes more and more apparent to me that this thing is going to come down to whoever gets the vote out more.

Fire up the vans and break out the walking around money.
 
2012-10-09 02:50:48 PM
Ya'll keep pointing to the electoral college as Obama's ace in the hole. And I admit, he's still in a very very strong position.

But elections are about trends, up until last Tuesday, Obama had Romney in a consistent downward trend. A trend that was all but ensuring his re-election.

Now, the trends are reversed and the question is: is this a bounce for Romney (indicating a temporary shift in public opinion) or is this lasting? And even more importantly, will it continue? Shockingly, Obama can afford to lose even Ohio, Florida, and Virginia so long as he keeps Nevada, Colorado, New Hampshire and Iowa. But, it becomes harder to hold on to Iowa, Colorado and possibly (though unlikely) even New Hampshire if he's losing those three swing states.

Quite frankly, I'm stunned at how reluctant the campaign has been to bring out the name and images of Bush II. I know that the Obama's appreciated how Bush and his family helped make the transition smooth and that the President has some respect for him as an individual. But you're fighting for the f**king country here, and if you were to poll American's about who's foreign, economic, and tax approach they preferred: Clinton vs Bush II, Clinton would win overwhelmingly. Obama is using Clinton's playbook, Romney is using Bush's. Why in the world wouldn't they be smashing Romney over the head with that fact over and over again? The hell with this Big Bird nonsense. Romney = Bush III.
 
2012-10-09 02:53:54 PM
This really wasn't that hard to predict. The less and less informed and attentive a populace gets, the more apt they are to react to instant stimuli instead of actually thinking about things.

People want to feel like they are participating in something important like the election, but they also want to rush home so they don't miss the latest episode of Honey Boo Boo. They hear that Romney won the debate and decide to vote for him, and feel like they have done their due diligence, even though they put about 30 seconds of energy or thought into it.

If Obama comes out in debate two and three and reduces Romney to the sniveling ball of lies that he is, they will do the same thing again for Obama.
 
2012-10-09 02:54:56 PM
Called my bookie, nothing has changed. You guys got me all excited, I wanted to win big on Obama like the last time around.
 
2012-10-09 02:55:48 PM

ManRay: Every day it becomes more and more apparent to me that this thing is going to come down to whoever gets the vote out more.

Fire up the vans and break out the walking around money.


Unfortunately, ACORN was really good at getting folks to the polls.

I honestly couldn't tell you what group to give your money to to help with that, if you can't do time. League of Women Voters is generally pretty awesome at registering people, but I don't think the local chapter here is doing getting-to-the-polls stuff... ACLU here has been pretty good in past years about providing decent voter education materials... but other than specific campaigns working on their own thing, yeah. No idea. Something I need to look into, I suppose.

Obama's campaign was really freaking awesome with that in 08, though. I assume most of the ground work stuff is just happening there now, since the at least 3 groups I can think of that were doing non-partisan (but aimed at low-income communities, so, you know) voter drives-to-polls and turnout work 6 years ago simply don't exist anymore.

We're doing GOTV stuff at work but the folks who support us aren't the ones that really need help/encouragement/education to get to the polls (education being ESPECIALLY important with OH and PA's back and forth on Voter ID...).
 
2012-10-09 02:59:49 PM

Citrate1007: The only poll that counts is the popular vote


Uh, no.
 
2012-10-09 03:00:02 PM
The same people who are too lazy to work are too lazy to vote. Take away those people from the poll and Obama's lead is nonexistent.
 
2012-10-09 03:01:17 PM
This was always going to happen. As long as Romney didn't shiat himself in the first debate, he was going to be declared the winner. The media would use the debate narrative and every other means possible to paint the election as being close.

After the next debate, the big story will be that Obama did much better, while Romney did well but didn't quite meet expectations, even if they give the same exact performances. The media will try to spin it as Obama making a comeback, but Romney still carrying the momentum in a very tight race.

Meanwhile, the polls will move a few points here and there, but the electoral math will not change very little if at all. Only a huge gamechanger can flip enough states to alter the fundamentals of the race, and it's looking more and more like there won't be one.
 
2012-10-09 03:03:10 PM
Maybe he just doesnt want the job anymore. I dont blame him, I wouldnt want to save this shiathole either.

Burn it all down 2012, vote Romney.
 
2012-10-09 03:04:01 PM

StreetlightInTheGhetto: //but I'm over my BS NYT quota
///if you search an article from them via Google News you can get around it, but it's stupid, so I'm done bothering


Does it apply to the 538 blog? I've never been prompted when visiting it.
 
2012-10-09 03:04:18 PM

keytronic: Now, the trends are reversed and the question is: is this a bounce for Romney (indicating a temporary shift in public opinion) or is this lasting


No. Because at the end of the day Romney is still Romney and a plurality of Americans at least are sane.
 
2012-10-09 03:04:53 PM

Pocket Ninja: In which "people who are actually voting in the election" actually means "people Gallup considers likely to vote in the election," which in turn means "a changing demographic that we can alter at will," which in turn means, "a target group malleable enough from sample to sample that we can consistently show results suggesting a narrow contest," which in turn means, "we stay marketable and our polling remains in high demand."


1.bp.blogspot.com

'yeah, you know, that's just like, uh... your opinion, man!.'
 
2012-10-09 03:06:02 PM

Altair: Called my bookie, nothing has changed. You guys got me all excited, I wanted to win big on Obama like the last time around.


People who bother to risk their own money on the election tend to actually pay attention to what is happening, rather than hunt for ways to make a compelling story. So when Gallup applies their LV screen and the numbers change, they don't freak out like news anchors and low information voters such as Subby.
 
2012-10-09 03:06:10 PM

ShawnDoc: StreetlightInTheGhetto: //but I'm over my BS NYT quota
///if you search an article from them via Google News you can get around it, but it's stupid, so I'm done bothering

Does it apply to the 538 blog? I've never been prompted when visiting it.


I don't think the blogs count, just the main site articles.
 
2012-10-09 03:07:36 PM

Citrate1007: The only poll that counts is the popular vote, and even that can be circumvented by a governor who's your brother and a SCOTUS that was put in place by your father and his friends.


i.qkme.me
 
2012-10-09 03:10:35 PM

Hollie Maea: Gallup's LV screen tends to be somewhat GOP leaning. Thus said, here is where we stand today:

Would Obama have lost if the election was held over this weekend, the time when the polls rolling in today were conducted? Most likely.
Would Obama lose if the election were today? Probably not, but it would be very close.
Will Obama lose in November? No, unless he farks up both of the next debates.


It's a setup for the comeback kid narrative.
 
2012-10-09 03:12:11 PM

Hollie Maea: Gallup's LV screen tends to be somewhat GOP leaning.


Older people are more consistent voters. Older voters tend to be more conservative, thus the bias you're seeing. It's pretty standard.
 
Displayed 50 of 86 comments

First | « | 1 | 2 | » | Last | Show all

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report