Senate bill supporter Nate Silver: Do not trust Firedoglake polls

[ohiodemocratic]

Nate Silver today lashes out against his nemesis, the Firedoglake website (which is against the Senate bill and in favor of the House health bill). Silver says that the SurveyUSA polls that Firedoglake endorses are misleading.

On the other hand, Silver has long held the view that Jonathan Gruber is to be trusted despite the humongous sum of money he receives from the entity he constantly praises.

Link here: http://www.fivethirtyeight.com/2010/02/can-you-trust-those-firedoglake-polls.html#comments

[Pirate Smile]

1. I trust Nate Silver a hell of a lot more then Hamsher. Easy choice.

[CBR]

3. Yup... nt

[berni_mccoy]

4. +1. Norquist is behind FDL now.

[ohiodemocratic]

7. That's false

Norquist and Hamsher wrote one (1) letter calling for an investigation of Rahm Emanuel's tenure in Fredie Mac circa 2001. Does that qualify as "running" things?

[JackRiddler]

24. Sounds reasonable.

Now they should call for an investigation of a quite a lot more!

[ohiodemocratic]

5. My condolences, regarding the death of LieberCare

It must have been a blow to Silver at. al.

[SemiCharmedQuark]

8. Absolutely.

[HughMoran]

16. +77

[grantcart]

55. +1

[Arkana]

2. Nate is a whole hell of a lot more trustworthy than Jane "I Heart Grover Norquist" Hamsher.

[redqueen]

10. Jane "I Heart Grover Norquist" Hamsher?

What's that about?

[ohiodemocratic]

13. Silver co-signed a petition with Grover Norquist. Nate Silver hearts Norquist!

This is truly a bipartisan endeavor, with everyone from Markos Moulitsas to Grover Norquist on board.You can sign our petition to Demand Question Time here, and follow us on twitter

http://www.fivethirtyeight.com/2010/02/few-questions-about-questiontime.html

[SemiCharmedQuark]

14. That was a letter signed by a multitude of people, not just two.

Hamsher's letter was just her and Norquist and was coauthored by both, not just signed by them as the petition you mention was just signed by a multitude of people including Kos, Silver, Norquist, etc.

[Orsino]

27. Don't get in the way of the pitchforks.

The Puritans are on the march.

[ohiodemocratic]

33. So you mean Silver hearts a lot of people INCLUDING Norquist?

The fact that Norquist is not the only person Nate hearts is irrelevant.
We're discussing whether Nate hearts him or not, aren't we?

[SemiCharmedQuark]

36. Did Nate Silver COAUTHOR a letter with Norquist? No.

These are the people who authored the letter:

Micah Sifry, Mike Moffo, David Corn, Mindy Finn, Jon Henke, Glenn Reynolds

And at that point, whether or not someone signs it is only up to them. It does not mean there is a partnership between those who sign it any more than I am in a partnership with the other people who signed a petition to keep a nearby library open.

[ohiodemocratic]

38. Feingold hearts McCain! They co-authored a bill once!

Your argument is inconsistent and a big hole has been poked in it.

1) Co-authoring a good thing is not bad. It all depends on what is being said in the letter, petition, or whatever type of document we're talking about. But you pretend that the content of a letter is irrelevant. It's childish to say that if two people simply co-author something, then one loves the other.

Do you doubt that Hamsher disagrees with Norquist on many things? Do you think Norquist wants a public option? Is one a sufficient number to conclude a person loves another?

This is simple: you do not like Hamsher because she has views that are different than yours regarding health care reform. That makes you pretend she loves Norquist.

[Arkana]

40. I know that they both share a healthy hatred of the President

and desire to see him fail.

[ohiodemocratic]

42. Um...McCain-Feingold was written before Obama was President

In 2002. Odd.

[Arkana]

43. Not referring to McCain and Feingold.

I am referring to Hamsher and Norquist.

[ohiodemocratic]

46. But you replied to a post in which I refuted the idea that co-authoring something is always bad

Why did you reply to that comment then, if you were not going to address the subject?

Is co-authoring with a right-winger is always bad?

[Arkana]

47. Read your own damn post.

That wasn't the only assertion you made.

I was addressing the portion of your post that was about Hamsher and Norquist. I don't give a shit if they DO disagree on everything else. They team up with the express purpose of bashing the President and making headlines.

[ohiodemocratic]

50. Get the content of the letter right: The purpose was to investigate Emanuel

And why don't you give a damn about the rest of Hamsher's views? Are you so narrow-minded that you focus on one thing?

[Arkana]

60. I don't give a damn about the rest of Jane Hamsher's views

because she is so holier-than-thou that she makes me want to vomit.

I agree with Dennis Kucinich on about 90% of policy too, but I don't particularly like him because he insists on grandstanding rather than getting shit done. You want to start with me on that too?

[SemiCharmedQuark]

66. McCain is less odious than Norquist

And McCain and Feingold shared a goal: Campaign Finance Reform

Norquist's goal is to see Democrats and Obama fail. Hamsher's goal is single payer. She believes that this tack will get her there by getting rid of Emmanual and Norquist just wants the Dems to fail.

[HughMoran]

17. +78

[Luminous Animal]

32. Nate never said what the OP claimed. The title of this OP is a lie.

Did you even read the article?

I swear, people on DU are like flying monkeys.

[WeDidIt]

6. Nate Silver called the 2008 election dead on

Jane Hamsher is a proven liar.

I know who to trust.

[ohiodemocratic]

11. Rasmussen is pretty accurate too n/t

[depakid]

19. i'm asuming that's sarcasm...

[Arkana]

22. Rasmussen called the popular vote percentages.

Even a stopped clock is right twice a day.

Nate not only called it, he called every state correct except Indiana and Missouri.

[ohiodemocratic]

34. Rasmussen was 24 for 24 in 2004 (and Silver is not a pollster. He only picks names)

Rasmussen has not been right in two or three polls, as you imply: http://www.rasmussenreports.com/public_content/politics/elections2/election_2004/state_by_state_actual_results_vs_rasmussen_reports_polls

Let's see Silver conduct polls and see what happens. He only picks winners based on his review of the universe of polls.

[Arkana]

41. ...

http://www.fivethirtyeight.com/2010/01/is-rasmussen-reports-biased.html

Nate's pretty good at what he does, and he shreds Scott Rasmussen and his Republican-friendly polls.

FYI: Rasmussen also predicted Hillary would crush Obama in the primaries, and that NY-23 was a lock for Doug Hoffman the day before the election.

[ohiodemocratic]

44. So what percentage of the time do you think Rasmussen has been right?

?

[Arkana]

48. I present to you this:

http://www.surveyusa.com/index.php/2008/02/06/2008-pollster-report-card-through-super-tuesday-active-pollsters-only/

Ninth place out of 14 active pollsters. Not real good.

[tritsofme]

59. lol, I might be suspect of ranking system that names itself as the winner

Reminds me of the Dick Cheney Method for VP selection.

And the pollster that put itself as first place in your link, is coincidentally the same pollster that Hamsher commissioned, cited in the OP. Now I have no particular problem with Survey USA, I just found your cite amusing.

Either way, reliably polling congressional districts is dubious for any pollster. Judging Rasmussen based on a NY-23 special election is just silly.

Rasmussen has had a sterling record in the state and national races since 2004. I don't care about the owner's politics, he produces consistently good results.

[ohiodemocratic]

49. Just curious: Any links to Rasmussen's opinion that Hoffman was "a lock"?

Or could it be that Rasmussen said no such thing?

[HughMoran]

18. +79

[Sebastian Doyle]

28. How much Diebold stock does he own?

Just askin'

[depakid]

9. Interesting- I'd like to see a lot more of this from Nate

A LOT of the cheap media polls (and other surveys) Nate's discussed have methodological problems that in the past, he's tended to often to gloss over.

[Spazito]

12. Yep, good advice by Silver...

FDL lost ALL credibility when it's founder joined hands with the likes of Schlafly and Norquist not to mention advocating joining with the teabaggers, imo.

[Luminous Animal]

51. Except Nate never said that. The OP is a lie. Did you even read Nate's article?

The title of Nate's article poses the question, "Can You Trust Those FireDogLake Polls?" Then he proceeds to critique the methods of the pollster SurveyUSA (the firm FireDogLake hires to conduct polls) for which he claims to respect and regard as "a strong and transparent polling firm."

He concludes that, "The last thing that needs to be said is that whatever numerical differences there are, they're on the shoulders of SurveyUSA and not FDL. In an e-mail exchange with me, SurveyUSA's Jay Leve made it quite clear that his firm was responsible for all methodological decisions about its polls."

So let's recap:
1) SurveyUSA is hired by FDL
2) Nate not only respects SurveyUSA but finds them a strong and transparent polling firm
3) Nate analyzes their methodology in one poll
4) He offers a measured critique of that methodology and concludes that he has some issues but
5) NOT ONCE DOES HE CONCLUDE OR EVEN INTIMATE THAT FDL POLLS CANNOT BE TRUSTED.

[Spazito]

67. FDL has NO credibility, imo...

It has become Nader central and it's founder has NO problem cozying up with the likes of Norquist, Schlafly and the teabaggers if it helps promote contempt for Democrats and the Obama administration.

Nate very clearly finds Survey USA methodology suspect, this is not the first time he has written on that subject and FDL uses Survey USA because the flawed methodology gets the results that further FDL's agenda so I don't buy your defense of the indefensible.

[RoyGBiv]

15. Silver's correct ...

Silver's math is good and observable. He explains his weighting process and logic.

And he just happens to be right most of the time.

[Luminous Animal]

52. What is he correct about?

[RoyGBiv]

57. Lots of things ...

In this particular case, he is correct in his critique of FDL's polling recent methodologies and strategies.

[Luminous Animal]

58. Except Nate never claimed that FDL's (Actually SurveyUSA) polls are untrustworthy.

And he spoke nothing of their strategies.

So, again, what specifically is he correct about?

[RoyGBiv]

62. That would seem to be incorrect ...

Note that I didn't say anything about "trustworthiness" and instead said that he is correct in his critique of the methodologies and strategies.

"As regular readers of this website will know, I have very little respect for FDL's 11-dimensional chess strategies."

I am a regular reader of his website, and I did know that, but as this sentence indicates, he did mention his lack of respect for their strategies in this piece.

"So I don't think there's anything untoward that's gone on here -- although there do appear to be some house effects in these congressional district polls resulting from methodological decisions that SurveyUSA has made."

[Luminous Animal]

63. Will honestly claim that you did not understand my point?

That being, in this article, Nate did not analyze FDL's strategies (he merely took a gratuitous swipe).

Nate analyzed SurveyUSA's methodologies. A polling firm that he respects and thinks is strong & transparent.

He never arrived any where close to the conclusion that SurveyUSA's polls were not to be trusted.

[RoyGBiv]

64. Gratuitous?

In what way was the "swipe" gratuitous? Was he not simply emphasizing a point he has made repeatedly and that does in fact seem to have bearing on the questions he is addressing?

Whether I understood your point is irrelevant. You asked a question that I answered. If you wish to make a point, please do so, and I will reply to that.

[FrenchieCat]

20. I'll take Silver when he speaks of polls

over anyone else. Period.

[Arkana]

23. I just hope he sticks to polls.

Chuck Todd was good when he stuck to numbers only--but once they started asking him for political opinions he got full of himself.

[Cosmocat]

26. PAYCHECKS and power ...

I swear, when I first saw Todd on C-Span in his hotline days, he was a progresssive ... I just got that feel, he seemed to enjoy breaking good news for Ds ...

As he has crawled up the corporate ladder, he has sold out completely, it has been fascinating to watch ...

NOT that I think people should be biased toward Ds, just call it fair and square.

But, the MYTH of the liberal media is that people know the only way to move up is to 1) play the Ds are in trouble meme 2) always leave the Rs with an out

MSNBC and the KO, Maddow, Big Ed, and Shuster and some of the other daytime people are exceptions ... But, the general rules to move up are 1 and 2 ...

[Luminous Animal]

31. Well that is good

given that he's offering a critique of SurveyUSA (the polling firm that FDL hires) which he regards as "a strong and transparent polling firm."

And that he says this:

"The last thing that needs to be said is that whatever numerical differences there are, they're on the shoulders of SurveyUSA and not FDL. In an e-mail exchange with me, SurveyUSA's Jay Leve made it quite clear that his firm was responsible for all methodological decisions about its polls."


Did you even read the frickin' article?

[MadHound]

21. I don't give a damn what either one of them says, nor am I in favor of either bill

Both are going to screw us, so I say start over and this time construct a truly progressive bill, not something that is faux progressive and designed to benefit the insurance industry.

[FrenchieCat]

25. Yeah, yeah.......

we know already! :boring:

[Lilith Velkor]

30. Ever wonder why your OPs about your personal life get unrecced to oblivion?

Maybe it's because all those nasty little digs add up.

[FrenchieCat]

35. Yeah...Like Madhound has ever anything but digs to dole out!.

Here Madhound is in the thread you mention.....
the thread about my 7 year old nephew whose stage 3-4 Cancer was diagnosted at age 18 months!
There is your "wonderful" Madhound saying the same damn thing in essence as he has just said in this thread!

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=433&topic_id=181217&mesg_id=181471

So pardon me for not being "nice", and thank you for your own very special concern about
what I post, and when it is that you decide something is soooo important enough to chime in,
and when you choose to stay mum! Yeah....it does add up, don't it?

And I'm sure you were most likely one of those unrec's.....
so I'm glad you feel all justified and shit. Goodie for you...
you've got yourself a real weapon against me a my 7 year old nephew! :eyes:

[Name removed]

65. Deleted message

Sub-thread removed by moderator. Click here to review the message board rules.

[Luminous Animal]

29. Nate Silver does not say "Do not trust Firedoglake polls"

Not once does not say this. He poses his own question then analyzes FDL's methodology part of which concludes this:

The last thing that needs to be said is that whatever numerical differences there are, they're on the shoulders of SurveyUSA and not FDL. In an e-mail exchange with me, SurveyUSA's Jay Leve made it quite clear that his firm was responsible for all methodological decisions about its polls.

...

But ... none of that necessarily impacts the accuracy or integrity of the horse race question, which SurveyUSA has signed off upon. And as little respect as I have for FDL is as much respect as I have for SurveyUSA, which is a strong and transparent polling firm.

A better more honest title to Nate's article (the title of this post is totally dishonest) would have been, "Can you trust those SurveyUSA polls?"

[t0dd]

45. +1.. the OP is deceiving readers here

[spiritual_gunfighter]

37. Dishonest OP if there ever was one

That isn't the title of the blog post, but you already know that.

[ohiodemocratic]

39. I didn't use quotation marks. Who said I'm directly quoting someone?

If John Doe says "abortion is bad" and I post a thread saying "John Doe opposes abortion," was I being dishonest? No. Just because I didn't use quotation marks doesn't mean I was being dishonest.

Silver concludes that Firedoglake polls are not trustworthy. You would have to read the whole story to realize that.

[Luminous Animal]

54. I did read the entire story and Nate made no such conclusion.

Quote it if you've got it. His critique of SurveyUSA's methodology (a firm for which he respects) is far more nuanced than that.

[spiritual_gunfighter]

61. I read it

But maybe you should follow DU rules next time.

[Unvanguard]

53. Did you actually read what he said? n/t

[Dr.Phool]

56. It depends on how the polls are conducted.

We paid to have both Rasmussen and SurveyUSA do polls for us.

The polls that Rasmussen conducted for us, were fairly accurate. BUT, we wrote the questions to be asked in the polls, because we wanted specific information. Generic polls I've seen them do also tend to slew pretty much to the right. It's all in the specific poll's methodology.

SurveyUSA tends to be less scientific, and not as well conducted. They're better at giving a running position on trends. I don't know what the methodology was that SurveyUSA used in the FDL polls.

Just a comment from someone who's used them both.

[WI_DEM]

68. I don't really trust Survey USA much either and I endorse the House Bill.