Occam's Razor

[cthulu2016]

A cause and effect is more parsimonious than three or eight interlocking conspiracy theories.

For instance, the polls on the Walker recall election were dead on perfect. The RCP average was within 0.1% of the final total.

Unfortunately, half of those polls had already been "debunked" here... worthless, meaningless, biased, etc..

Using the "real" polls, Walker was going down. But he won. Since he won despite the fact that all the polls that hadn't been "debunked" said he would lose that was proof of fraud at the polls.

So we had unusually reliable polling (in aggregate) and a result that should have surprised no one and somehow that very boring reality becomes a multi-layered conspiracy.

The polls matched so well because "they" fake the polls and then "they" fake the elections to match the fake polls.

Yes, this is all possible... it could happen that way.

But if the polls are all fake and the elections are fake, then why vote at all? Why not do something productive, like betting on professional wrestling?

I don't think these constant conspiracy theories are the least bit helpful. Given that turnout, as a general rule, benefits Democrats I honestly do not understand the utility of the "your vote will not be counted, it's all fixed" mind-set.

A 6% lead is a substantial lead, but it represents only 3% of the population changing their mind one way or another. And there is 3% of the electorate in any state that swings like a weather vane. Elections are somewhat volatile.

If you have what polls as the worst debate performance in presidential debate history it will have an effect.

Subsequent positive develops can have an opposite effect. A jobs report. A good follow-up debate performance. Effective ads. A lot of donations.

Obama is still a favorite to win and will probably rebound somewhat in the future, but it is normal and expected that when 65 million people watch you look terrible it has some negative effect.

The race today is probably roughly tied among likely voters. It is an unfortunate fact that Democratic leaners are not the the likliest voters. If everyone voted we would seldom lose an election, but not everyone does.

Hopefully two weeks from now Obama will be up a few points with likely voters. I expect him to be. And polls will reflect those changes, as they have been doing all through the process.

After I wrote this I read this from Nate Silver that I think puts this well:

Polling data is often very noisy, and not all polls use equally rigorous methodology. But the polls, as a whole, remain consistent with the idea that they may end up settling where they were before the conventions, with Mr. Obama ahead by about two points. Such an outcome would be in line with what history and the fundamentals of the economy would lead you to expect.

http://fivethirtyeight.blogs.nytimes.com/2012/10/08/after-conventions-follow-the-bouncing-poll-numbers/

[Ztolkins]

1. Why are you writing this kind of...stuff.

Measured and whatnot.

I'll lend you my lighter, please use it to light your hair on fire. Thanks.

[11 Bravo]

2. No shit! Nuance and reason at DU? What the fuck is up with that?

[onenote]

3. why are you afraid of this particular analysis

If you want to rebut it, you should. But your post simply conjures up the image of someone putting their hands over their ears and going "naah naah naah..I can't hear you...naah... naah...naah".

My view of the current situation is that Obama did not do well during the debate, but its not the end of the world nor are the polls that are coming out this week. Indeed, they are totally predictable. Even without the debate, this race was almost certain to "tighten" up at some time during October. That's just the way these things tend to go.

[Ztolkins]

11. Hmm, I really don't think my post does that.

I pretty much agree with yours.

Maybe I should've said that I was being sarcastic. My bad.

[bhikkhu]

9. Unfortunately, "hair on fire" translates directly to "very likely to vote"

...and the rhetoric all around is skewed towards setting fires. I was raised to believe that voting was a duty and a right of passage (as was being informed about open-minded and all those good things), so you couldn't stop me from voting.

I don't understand why we have a Columbus Day and a President's Day holiday, but no Voting Day; like it doesn't really matter?

[porphyrian]

4. . n/t

[Blaukraut]

5. How many election cycles of unmotivated Dems is it going to take before they finally get it?

This is what pisses me off. "Oh yea, I'm a Democrat. I like Obama. I want equal pay. I want health insurance. I want medicare and social security kept safe....when is the election? ....oh, I dunno if i can make it to vote......"

[DemocratSinceBirth]

6. Three Points

1) Righties insist, incorrectly, that the polls showed Walker going down when he wasn't and therefore polls consequently can't be trusted.

2) I'm not a poll truther. I just believe Pew was in the field during a very volatile period and caught Romney's bounce and not his ebb.

3) We are in agreement. As a logical positivist Lord Occam is my hero.

[cthulu2016]

8. I have no opinion of the PEW poll

They are usually one of the Dem-friendlier pollsters so I don't have a bias objection, but any poll can be off.

I suspect the same poll done a day later might have been different.

Snapshots in time, and this is a volatile week.


As to what righties insist... well, that's life. They insist all sorts of things.

[grantcart]

7. Not sure what you are actually trying to say

But it is not a conspiracy to observe that there are vast differences in polling standards and their results.

http://www.realclearpolitics.com/epolls/2012/president/co/colorado_romney_vs_obama-2023.html

Colorado the U of Denver is showing Obama up by 4

and Gravis Marketing that has no established client base, has never participated in political polling before a a couple of months ago, does not have a sustainable business model, boasts no known political scientist, doesn't archive its results . . .

Shows Romney up by 3

[DemocratSinceBirth]

10. You Don't Have To Be A Poll Truther To Conclude Doug Kaplan Is A Crappy And Suspect Pollster

I think the OP is addressing the inclination to dismiss any poll whose results we don't like.

People are in an uproar over Pew. I just think if they were in the field at a volatile moment.

[cthulu2016]

12. A fake poll is a fake poll by definition

The gravis poll appears to be a fake poll, or so flawed as to be meaningless.

There are some polls that are absurd.

There are other polls that have unpopular results but are not facially dismissible unless one is rigorous enough to dismiss unduly popular results.

Once one starts cherry-picking the whole exercise becomes silly.

But a poll from out of nowhere that shows Romney winning blacks can be disregarded safely enough. The Gravis poll is silly.

And I would say the same about a national poll that showed Romney getting 40% of the white vote... it would be so far out of line as to be absurd.

But a poll showing Romney tied among registered voters (as the PEW poll does) is not facially absurd. It may be a flawed poll, but if we dig into the flaws of unpopular polls while uncritically accepting popular results then we cannot possibly have a sensible view of the world.

Scrutiny is good. One-sided scrutiny is not.

[grantcart]

13. Certainly agree that cherry picking isn't helpful.

But what recent events have showed, IMHO, is that there is no gate keeper that is doing due deligence on the pollsters or their methodolgy.

Right now recent DeVry graduate Doug Kaplan is THE MOST QUOTED POLLSTER on the web. Google Gravis Marketing and you can go 50 pages deep. The right wing bloggers and the Republican Party has made this guy more popular than Rasmussen.

As long as the numbers appear to match a certain pattern they are assumed to be right and added to the mix.

http://www.site.dougkaplan.net/Home.html

http://www.site.dougkaplan.net/Doug_Kaplan.html

Look at the graph for Colorado. Rasmussen, Gravis and the American Conservative Union turned the numbers upside down:

http://www.realclearpolitics.com/epolls/2012/president/co/colorado_romney_vs_obama-2023.html

My belief is that with a drastic drop in fund raising the Romney campaign was desperate for some revised figures to get some money into the campaign.