Romney, Obama, Rasmussen

[RandFan]

RCP Obama +.01

 

[RandFan]

This is actually humorous; I've been pointed at a couple of conservative blogs that refuse to acknowledge that is wrong, and rather insist that Jeep will in fact be moving American jobs to China.

Yeah, I have friends and family members who know that it is true. If it weren't then Romney would have owned up to it.

 

[332nd]

Not a surprise though, given that:
- Hillary had to beg him to stop dilly-dallying on Bin Laden and authorize a strike

Does it sting so much that Obama succeeded where Bush turned tail that you have to make stuff up?

- While our people got butchered in Benghazi, he apparently just sat there, frightened and paralyzed, then made sure to get some sleep before cruising off to a Las Vegas fundraiser.

I guess it does.

Too bad Dems, you should've nominated Hillary. You would've been waltzing towards a cakewalk re-election and we all would've had a president with some balls.

You can't watch Blazing Saddles any more without picturing Barack in place of Bart when Lilly screams "Its twue! Its twue!" How sad...

 

[A'isha]

Rumor, innuendo and lies in a new book. You can read all about it here.

Oh, that book. No wonder I didn't recognize what he was referring to...the Hillary Clinton thing wasn't mentioned back when the right first started trying to pass off its nonsense back in July.

 

[Buckaroo]

A guy so incompetent he can't beat an incumbent during hard economic times. That takes a real douche-bag.

Let's not get too cocky. 1 in 5 odds are still too high for my taste.

 

[RandFan]

Let's not get too cocky. 1 in 5 odds are still too high for my taste.

If Romney wins I can honestly say he won ugly. Seriously. He should have been consistently ahead. Instead he has been consistently behind playing catchup and dealing with gaffes. A little honesty here from my GOP friends, how many news cycles has Romney won since becoming the presumptive nominee? Very, very few, right?

 

[Kaosium]

RCP Obama +.01

Here's an interesting article on why there might be two different projected realities in this election and (perhaps) only one is likely to be right, thus splitting them down the middle (as RCP does) is probably just adding in delusion as opposed to rounding out reality.

The manifestation of these disagreements is evident in polling weights. Most Republican pollsters are using something close to a 2008 turnout model, with the same percentage of white, black, and Hispanic voters as the electorate that first elected Obama. Most Democratic pollsters are a little more bullish on minority turnout, which helps explain some of the difference between the two sides.

Add in a population that's changing its habits and pollsters have to contend with additional confusing factors. The number of Americans without landline phones is growing, particularly among younger voters. Those voters are much more difficult to convince to complete a poll, surveyors say.

What concerns Republicans most is the fact that media polls seem to track more closely with Democratic internals than with the GOP's numbers. Internal surveys conducted for Republican candidates like George Allen in Virginia, Richard Mourdock in Indiana, and Josh Mandel in Ohio draw much rosier conclusions than polls conducted for their Democratic counterparts Tim Kaine, Rep. Joe Donnelly, and Sen. Sherrod Brown. And media surveys, at least in Virginia and Ohio, show Kaine and Brown winning (restrictive Indiana laws make polling prohibitively expensive there).

Republicans say their party is a victim of media bias—but not in the standard "Lamestream Media" sort of way. Pollsters on both sides try to persuade public surveyors that their voter-turnout models are more accurate reflections of what's going to happen on Election Day. This year, GOP pollsters and strategists believe those nonpartisan pollsters are adopting Democratic turnout models en masse.

 

[Foolmewunz]

On the other hand here is the 'poll' the WSJ chose to...influence its headline today.

This poll is garbage, and in fact the WSJ relying on it might even depress Dem turnout, as it pretends that Obama is leading by eight. That might make some less likely to go to the polls as they figure the candidate they would've voted for will win easily without them bothering. This one the partisan ID is roughly what the Wisconsin electorate has been (over the course of many recent cycles) however it's not of 'likely voters' in fact it's not even of registered voters, it's a general population poll. Mind you in Wisconsin one can register the day of the election at the polls, but that's not exactly the best way to be determining between likely and less likely voters. Now here's the trump card, look at Question 2 on page 2:



90% responded they were absolutely certain to vote, 10% said they already had. Not a single one of the 1243 people they reached said they were 'very likely,' 50-50 or they didn't think they would vote. There's something obviously wrong with that, only about 70% of Wisconsin voters ever turn out, so a poll showing 100% of 1243 respondents saying that suggests something very strange must be going on. Marquette also polled for the recall, and they got much less bizarre results from that poll, 87% said they were absolutely certain to vote, although the actual figure was more along the lines of 60%.

How that occurred I couldn't tell you, it might not mean anything untoward regarding the pollsters, other than the fact they didn't notice that every single one of their respondents said they would vote or already had. It's not unusual at all, in fact it's expected, that many respondents might 'exaggerate' their likelihood of voting, but to get a result as astounding as that I cannot but suspect there might be another factor at work. One might be that Wisconsin has seen so much politicking--especially over the telephone--that many people simply won't respond at all anymore. The ones they do get aren't actually representative of the population, notably the independents which in this one must have skewed towards Obama against the grain of the rest of the nation; however that's not unlikely in WI either.

I dunno, but such a response at variance with reality ought to have caused someone at Marquette or the WSJ to wonder about the poll, as 1243 people all saying they were absolutely certain to vote or already had is basically impossible. Their sample pool may be polluted, the people doing the poll might be gundecking it--who knows, but what they should have known is getting 1243 people to all say they were certain to vote is basically an impossibility to achieve, despite the propensity of poll respondents to say they will when they won't. One way just occurred to me: perhaps they just polled those who'd already answered their poll with absolutely certain? They'd get a few who'd voted since early voting started, but the rest would probably give their previous answer. However they would also mean they were sampling their old samples and not getting anything random, which is not the best methodology in the business, as you don't want previous results polluting what is supposed to be a random sample.

I don't find that as surprising as you do. They whittle down their call lists for just this reason and try to only be calling the LV people if they want a valid poll. This seems to be a live call poll. Other polls head up their front page with 'likely voters only'. They may be throwing out the "I won't vote" responses so that they're not counting opinionated idiots who just want to get added into a poll.

Marquette polls trend left. For that reason, I agree that if I were the WSJ I wouldn't use it as an anchor to an article on the WI election landscape any more than I'd use Rasmussen's poll that trends in the opposite direction.

 

[Kaosium]

If Romney wins I can honestly say he won ugly. Seriously. He should have been consistently ahead. Instead he has been consistently behind playing catchup and dealing with gaffes. A little honesty here from my GOP friends, how many news cycles has Romney won since becoming the presumptive nominee? Very, very few, right?

I agree, but often that's driven by the reporting of it and this election is the worst for the media in my lifetime, there might be consequences for that, last time I recall it being this bad ('94-'96) led to Fox News being inflicted upon the world!

It has not always worked in the favor of Obama however, the first debate showed that: in my view it wasn't that Obama looked that bad (though he didn't have a good night) but that Romney looked especially good, notably in respect to the diminished expectations that most probably had for him. Before that his campaign was only described in terms of what 'gaffe' it had made, when in fact that was driven just by (many in) the media reporting whatever the Twitterazzi had determined would be the daily gaffe.

However the Dems put on a significantly better convention than the GOP, especially the last night, they ought to be proud of it, it was their best moment all election cycle.

 

[Puppycow]

I think that's the point he was making. The progressive - lib - union side in Wisconsin was very upset that Obama didn't use his bully pulpit to weigh in on their side or commit any of his considerable "personal" resources, e.g. the vaunted Obama ground team.

Mikedenk is, of course, hoping that Dems protest this by voting 3rd party (giving Rom/Ry the state).

Yes, and I hope that "real" conservatives in Virginia protest Mitt Romney's moderate positions by voting for Virgil Goode, thereby giving Obama/Biden the state.

 

[SkepticalDrew]

Nate Silver did an excellent analysis on my own state of Nevada. He's absolutely right, I'm not sure the political machine Reid and Obama built here can be beaten.

 

[Kaosium]

I don't find that as surprising as you do. They whittle down their call lists for just this reason and try to only be calling the LV people if they want a valid poll. This seems to be a live call poll. Other polls head up their front page with 'likely voters only'. They may be throwing out the "I won't vote" responses so that they're not counting opinionated idiots who just want to get added into a poll.

Does this mean I could end my agony (or at least diminish it) by just telling the next five pollsters I never vote? You wouldn't believe what it has been like here, in all the time I lived in WI I was polled (for politics) maybe a half-dozen times. I cannot tell for sure which of the ones I refused to answer as they came up with a suspicious caller ID were definitely pollsters, but I do know the robo-callers generally leave messages on my machine. I would guess that I've gotten three or so pollsters a week, or perhaps every two weeks, for about two months, with lesser but still significant amounts of polling (or attempts) since they decided to string up recall Walker.

However I disagree that's a solid polling methodology. That would 'lock in' past results and not sample the current population. There's a significant danger that those 'absolutely certain' they will vote will be less likely to change their minds, and of those who say they are 'probably' going to vote, most probably will, and not sampling them is quite likely to introduce errors on top of old data in my opinion.

Marquette polls trend left. For that reason, I agree that if I were the WSJ I wouldn't use it as an anchor to an article on the WI election landscape any more than I'd use Rasmussen's poll that trends in the opposite direction.

Legitimate polls (at least for projecting election results) shouldn't have anyone even wondering if they lean one way or another, they should be trying to hit the bulls-eye, which is the final result. That in aggregate Rasmussen has edged towards GOP candidates is meaningless as he's a reputable pollster and the odds are extreme he'd be on one side of the perfect result regardless. There's no reason to think that is a predictor of future events, as if he's off (significantly) in favor of GOP candidates that's not to his credit either, and the professionals try to adjust for that. He looks like a professional to me, someone who's trying to make a reputation with accurate polling, (some of) his detractors look like they have another agenda to me.

My guess is Rasmussen will be within the margin of error again, and it might be pro-Dem, and it might be pro-GOP. If it's true that there's a lot out there who appear to be accepting Dem assumptions on the electorate, that just means they're gonna be (more) in error about the final result. Your mileage may vary, and as it's just five days away now I guess we'll see who's right.

 

[MattusMaximus]

Predictwise...

Obama 70.0%
Romney 30.3%

I expect this weekend there will be some major selling of Romney shares at Intrade as people accept the inevitable. Already Romney is in free fall at Betfair and IEM.

I think it's all over now but the crying.

 

[DavidJames]

It will be interesting to see how the labor report tomorrow impacts everything. I expect it will dominate the news. If the rate goes up I also expect the (damn liberal) media to go nuts pontificating over the impact.

 

[Foolmewunz]

Meanwhile, back at Fox Opinion (I do like to separate their News which is simply biased in their choice of content rather than the actual reporting,... somewhat) that Hannity thing had on a grand echo chamber of both Morris and Rove today. The three of them were singing their big hit from the GOP convention "Oh, It's Gonna Be a Landslide", and talking the talk on Rove's whiteboard lies as to how the Dems are losing the early vote.

Would it be evil if I was to suggest that Sean should look to Howard Beale as a role model? I may make it through the election season without breaking my flatscreen, but my fascination with Fox is ending. I find myself turning them off more and more often.

 

[Kaosium]

Meanwhile, back at Fox Opinion (I do like to separate their News which is simply biased in their choice of content rather than the actual reporting,... somewhat) that Hannity thing had on a grand echo chamber of both Morris and Rove today. The three of them were singing their big hit from the GOP convention "Oh, It's Gonna Be a Landslide", and talking the talk on Rove's whiteboard lies as to how the Dems are losing the early vote.

Was Rove actually talking landslide? I just posted a column he did from the Wall Street Journal that had him predicting a three point victory with 279 in the EC, possibly more.

BTW, have you peeked at the demographic data of any of these state polls that are especially good for Obama? I just got done looking at Nevada, take a look at the non-Rasmussen polling demographics, if demographics mean anything and that electorate of 34-40% minorities, (up to) 10% more Dems than Pubbies with less than 10% Tea Party identification doesn't show up at the polls, then those polls aren't predicting the election result very well, are they?

Would it be evil if I was to suggest that Sean should look to Howard Beale as a role model? I may make it through the election season without breaking my flatscreen, but my fascination with Fox is ending. I find myself turning them off more and more often.

Heh, I find myself thinking of that movie more and more. It was almost prophetic---except it underestimated our appetite for sleaze. The other movie I find myself thinking frequently of is 'Wag the Dog.'

I dunno why you watch it, the whole presentation is garish. Shepherd Smith reminds me of an Amway salesman, makes my skin crawl!

 

[remirol]

if[/url] demographics mean anything and that electorate of 34-40% minorities, (up to) 10% more Dems than Pubbies with less than 10% Tea Party identification doesn't show up at the polls, then those polls aren't predicting the election result very well, are they?

Unskewedpolls.com is ---> that way.

Seriously. At this point, Romney's last chance is that _all_ the polls are somehow broken, and are consistently heavily skewed towards Obama. I rate the possibility of that as roughly 2%, tops.

It's far, far more likely that (as usual) some are tilted left to one degree or another, some are tilted right, and we really aren't going to have much of an idea whose methodology was 'best' until post-election. The pollsters simply aren't _that_ incompetent, taken as a group.

In 2004, RCP's average was Bush +1.5%; he won by 2.4%.
In 2008, RCP's average was Obama +7.6; he won by 7.3%.
In 2012 there's a noticeable split between national polls and the EC, but there's no reason to believe that RCP's straight averaging of all the polls is necessarily inaccurate; RCP has Obama +0.1 now and that's probably very close to true. Except that the Pubs have been picking up votes in hard-red or hard-blue states, where they don't matter, and Obama has the EC locked up.

 

[maxpower1227]

Oh my. unhingedpolls.com has a hilarious electoral map.

 

[Kaosium]

Unskewedpolls.com is ---> that way.

I came across that site at some point a while back but stopped reading when I saw him 're-weighting' for both Party ID and ideology (liberal versus conservative) you can't do that, for what should be obvious reasons.

However that does not mean that the concepts of weighting and demographics are of no value, and that they can skew polls if done poorly. I find it...surprising...that there would be anyone out there pretending differently. Note also that part of the GOP advantage with independents is diminished by assuming there's more identified GOP voters, especially if the electorate partisan ID figures return to the '04 neighborhood like Gallup and Rasmussen are finding.

Seriously. At this point, Romney's last chance is that _all_ the polls are somehow broken, and are consistently heavily skewed towards Obama. I rate the possibility of that as roughly 2%, tops.

Not all of them, and they don't have to be heavily skewed at all. The difference between a 'registered voter' and 'likely voter' poll was considered to be around 3-5% pro Dem, if there's polling firms out there employing methodology that simply makes assumptions about the electorate not likely to come to pass, which might very well have been pursued by someone part of the Obama campaign like Axelrod, that could easily turn those 'likely voter' results into (the equivalent of) 'registered voter' results. Now mentally move each poll you see that's not Rasmussen or Gallup 4% in Romney or other GOP candidates' direction and see what those electoral college maps look like.

It's far, far more likely that (as usual) some are tilted left to one degree or another, some are tilted right, and we really aren't going to have much of an idea whose methodology was 'best' until post-election. The pollsters simply aren't _that_ incompetent, taken as a group.

Mistaken assumptions by institutions leads to stock market crashes, financial meltdowns and industries requiring government bailouts, it's hardly unimaginable a handful of polling firms could talk themselves (or be talked) into adopting a turnout model that assumes that the '08 election made for a sea-change in American politics and would be built upon instead of regressing back to the mean like it usually does. The '10 elections and the fact that even with partisan ID numbers from some of the firms averaging 6-8 points pro-dem the election is still close ought to suggest there hasn't been any permanent re-alignment in American politics.

Also keep in mind how we know of (most) these polls: they're commissioned by papers and other media outlets. All they have to do is tell the polling firm they want certain parameters for the polls(s) they're purchasing, and if Axelrod will contact Gallup (!) and tell them their methodology is 'faulty' do you suppose he's failed to already let the (important) media know just what they should be 'expecting' in the '12 election and to be sure to 'demand' it from the polls they commission? What you might call 'conspiracy' is called strategy in politics...

In 2004, RCP's average was Bush +1.5%; he won by 2.4%.
In 2008, RCP's average was Obama +7.6; he won by 7.3%.

My guess is this election cycle will be different.

In 2012 there's a noticeable split between national polls and the EC, but there's no reason to believe that RCP's straight averaging of all the polls is necessarily inaccurate; RCP has Obama +0.1 now and that's probably very close to true. Except that the Pubs have been picking up votes in hard-red or hard-blue states, where they don't matter, and Obama has the EC locked up.

Not even Nate Silver is suggesting it's 'locked up.' Oddly enough he used an analogy I used in one of my posts before the conventions when I first started posting on this subject here. He says the odds are the same as a NFL team with a three point lead with three minutes left. That sound like a 'lock' even to you? I suspect he's not reading the scoreboard correctly and they're really down by a TD--and the other team has the ball. Go ahead and go through those state polls and move the ones not Gallup or Rasmussen four points in Romney's favor and see just how that changes the picture.

Anyway, I wouldn't want you to be farmed by any predatory capitalists in the penny-stock political market if you happen to be playing this year, so I figure I'd suggest that you get out while you can having made a tidy profit at any rate.


Maybe I'm wrong, but I honestly think this election will make a fair number of these fly-by-night pollsters look silly the day after. Just take a look at a wide range of those state polling demographic data and ask yourself if the electorate doesn't turn out like that whether or not you'd expect to get the same result that the poll is suggesting. Not just one, but look at as many of them as you can...

 

[A'isha]

My guess is this election cycle will be different.

Why?

Go ahead and go through those state polls and move the ones not Gallup or Rasmussen four points in Romney's favor and see just how that changes the picture.

Yes, if you alter all the poll results to give Romney four more points than those poll results say he actually got, it totally changes the game and gives Romney the advantage in polling.