NATE: Consensus Points to 50-Seat GOP Gain in House, But May Understate Uncertainty (20 to 80 seats)

[jefferson_dem]

Consensus Points to 50-Seat G.O.P. Gain in House, But May Understate Uncertainty
By NATE SILVER

FiveThirtyEight’s projection for the U.S. House shows little change from last week. Republicans are given a 73 percent chance of taking over the House, up incrementally from 72 percent last week. During an average simulation run, Republicans finished with 227 seats, up from 226 last week; this would suggest a net gain of 48 seats from the 179 they hold currently.

However, there is considerable uncertainty in the forecast because of the unusually large number of House seats now in play. A gain of as large as 70-80 seats is not completely out of the question if everything broke right for Republicans. Conversely, if Democrats managed to see a material rebound in their national standing over the final two weeks of the campaign, they could lose as few as 20-30 seats, as relatively few individual districts are certain pickups for Republicans.

In past weeks, we have written about the divergence between the various indicators that the model uses — for instance, the generic ballot, as compared against polls of individual districts. Increasingly, however, these metrics are falling into alignment.

Some generic ballot polls have shown incremental improvement in Democrats standing — although they still trail by roughly 6 points among likely voters on the generic ballot, according to our model’s estimate. According to one commonly-used formula, a Republican lead of 6 points on the generic ballot would translate into a gain of about 50 seats.

<SNIP>

http://fivethirtyeight.blogs.nytimes.com/2010/10/16/consensus-points-to-50-seat-g-o-p-gain-in-house-but-may-understate-uncertainty/

[Gman]

1. A spread of 60 seats and this is supposed to be credible???

Again, the methodology used is absolutely stupid and this means nothing.

[DemocratSinceBirth]

3. Sabato Has The Dems Losing 47 House Seats, Eight Senate Seats, And Seven Governorships

He was spot on in 06 and 08.

[Unvanguard]

11. How does that make any sense?

If the uncertainty is a reflection of actual political conditions, it is a feature, not a bug.

[DemocratSinceBirth]

2. This Is Simple.

The economy sucks. Folks in their desperation and despair are looking for someone to blame. Most of them don't take the time to determine if the blame is justified or not. The Democrats are in power. Therefore they will be blamed.

If the economy doesn't improve by 2012 we will be in the same position but the stakes will be even higher.

[iamjoy]

7. Post hoc ergo propter hoc

Conservatives have conned the American people into believing that because the dawn came after the rooster crowed, the first event caused the latter.

[DemocratSinceBirth]

8. It's A Logical Fallacy

However it works both ways. We were the beneficiaries of a lousy economy in 08. Let's hope it improves by 010.

[golfguru]

29. You are wrong about 2012

From Jan 2011 through Nov 2012, republicans will be in charge
of congress and can be blamed for the horrible economy. Obama
will win in a cake walk.

[DemocratSinceBirth]

31. The Incumbent President And His Party Is Going To Be Held Responsible For The Economy

It has been the case in every election. He is the focal point for everything . That's why Bush Pere lost in 92 despite having a Democratic Congress and Clinton won in 96 despite having a Republican Congress.

This would be true regardless of who is in the White House.

Do you really think Democrats will have a good year if unemployment is still hovering around ten percent?

[golfguru]

33. Yes, even if unemployment is around 10%

Obama will have enormous ammunition to blame on the
do nothing, social security & medicare cutting repubs.
In 2010 mid-terms, dems are facing all the bullets because
everything is majority dems.

As for the Bush41 loss in 1992, chalk it up to the brilliant
politician, Bill Clinton and the read my lips comment.

[DemocratSinceBirth]

37. If A President Of Either Party Can Win Reelection With Nearly Ten Percent Unemployment

It stands everything we know about voting behavior on its head.

[golfguru]

38. There is first time for every thing!

You heard it here first.

[impik]

4. So basically no one knows shit

[bigdarryl]

5. impilk your 110% they don't know shit I have more respect for Congressman Clybourn than these idiots

he's saying he thinks we will loose seats but still retain the congress.He also said on Ed's show that there will be surprises in NC and that across the country African Americans are energized about this election

[jefferson_dem]

10. Clyburn is a politician. It's his job to set expectations, spin, and frame electoral politics.

But Nate is an idiot cause he's taken a hard look at the data, done some most sophisticated number crunching, and admits that things are very much in flux and election outcomes are unpredictable. Right-o. :crazy:

[Imajika]

13. You realize he pretty much has to say that though right?

Not saying he will prove to be wrong, just that elected Democrats saying Democrats will do better than expected is kind of part of their job.

Polling does look pretty bad, but if we are successful at turning out our base our losses may be minimized. Just holding the House and Senate would be considered a victory - though that still allows for the Republicans gaining 38 seats in the House and 9 in the Senate. We have so many more seats than the Republicans that even an even voter split would result in a lot of gains for them.

[monmouth]

6. And covering their a$$es...n/t

[jefferson_dem]

9. The only thing we know is...

that we don't know. As a social scientist, I appreciate that Nate just comes right out and concedes that ... nobody really knows ... and this electoral climate is even more volatile than in other years. Political partisanship aside, November 2-3 will be fascinating to watch unfold.

[Unvanguard]

12. No, the centerpoint of the spread is valuable knowledge

especially because the probabilities around it are normally distributed, so we know the end points are far less likely than something reasonably close to the present estimate of +50.

What is means is that the Republicans have solid but not yet certain chances of taking the House, but are unlikely to take it by a strong majority.

[jefferson_dem]

15. I'm not so sure. I think Nate's analyses indicate that...

median probability is 50 seats but the spread is vast - thirty points on either side of the median -- from 20 to 80. So the Repubs could fall easily short of the 39 or gain a fairly substantial majority. :scared:

Not to get too wonky but the predictions reflect a wider range (kurtosis)... More like A or C than B below.

[Unvanguard]

16. I don't think we're disagreeing, except maybe rhetorically.

By "unlikely" I did not mean "impossible", and by "strong" I meant something like what the Democrats enjoy now, which would be at the very high end of the uncertainty interval.

[pstokely]

39. So the House remains a toss up?

nt

[Kdillard]

44. I expect Dems will lose some seats and we will see where they fall based on Nate's predictions in

Nov. It could either be an epic night for Republicans or they could fall short. I think we need to be worrying less about Nate and more about voting and calling people to encourage them to vote. The rest will take care of itself.

[onehandle]

14. Consensus = 50. The consensus is more often closer than any wild card.

[Mass]

18. Except consensus is often a way to CYA. If you have the same prediction than the others,

you wont be attacked if you're wrong. (See NH's primary, for example).

[Mass]

17. Otherwise said, we dont know, except for knowing the GOP will win seats.

No surprise. I've known that for weeks now.

[Aramchek]

19. The more Nate talks, the less intelligent he appears.

[DemocratSinceBirth]

20. What If He's Right ?

Larry Sabato just said the Dems will lose approximately fifty House seats. Not only did he nail 06 and 08 but he teaches in one of the most prestigious political science programs in the nation.

[Aramchek]

22. will you still defend his predicting skills if he's not?

[DemocratSinceBirth]

25. Sabato And Silver Are Predicting Fifty

I'd say it's a reasonable bet.

[Upton]

21. Would you be saying that if Nate was predicting a Democratic win?

C'mon now, tell the truth..

[Aramchek]

23. Nate knows he is using faulty polling methods, yet still sells them as factual

It's not the result I necessarily reject (even though it's a crappy result),
it's the way he gets to that result.
And the things he says besides the numbers are what really show his intent.

[Upton]

27. Nate got 49 out of 50 states right in '08..

He predicted a Democratic victory right up until election day. Nobody was complaining about his methodology here then. In fact he was celebrated. I see you weren't a member of DU at the time, but I doubt if you would have had issues with Nate in '08 either.

And what exactly do you mean by that last comment? Are you accusing Nate Silver of intentionally lying about the current state of the upcoming election?

[DemocratSinceBirth]

28. Nate Said He's An Idependent Who Leans Democrat

There's no reason to disbelieve him. We look as foolish as the yahoos at Freak Republic who questioned the lions share of polling information that showed them getting waxed in 06 and 08.

[Upton]

30. Exactly

"the polls are BS"... "the media's against us"...it's all the same garbage. The only thing that's changed is the party ID of the folks doing the whining. I think what we're seeing is a lot of people in different stages of denial.

[DemocratSinceBirth]

32. Economy Is In The Ditch. Folks Are Looking For Somebody To Blame Even If The Blame Is Unwarranted

If McInsane was in the White House it would be the Republicants who would be getting waxed.

Everything else is window dressing.

[Blue_In_AK]

42. Maybe he got it right in the presidentail race,

but he blew it here in the House race. I don't believe anyone's polls in Alaska. It's too hard to get accurate results when over 60% of the electorate are nonpartisan.

[jefferson_dem]

35. What "faulty polling methods" is Nate using?

Please be specific.

[DemocratSinceBirth]

26. Nate And Sabato Nailed 06 And 08

They deserve the presumption of competence until they prove otherwise.

[old mark]

24. I rec'd this because I believe it is a fine example of the lies the media and the GOP

are telling themselves-and us. They are trying to discourage Democrats and those few "undecided" voters from voting. I believe much of the 'publican campaign donations is channeled to media payoffs, from the TV networks to newspapers.
Statistcs-polls-can be made to support any finding depending on how they are constructed, and if they in fact take place at all.

Silver has sonsistantly stated that Democrats will lose, and he is not changing his story. Why bother to even read his bullshit?

THE ONLY POLL THAT MATTERS IS ON NOVEMBER 2nd!!!!
And that one is up to the Democrats to win or lose.

mark

[Blue_In_AK]

43. Thank you, Old Mark.

You reinforce my belief that polls are voter suppression tools.

[BootinUp]

34. This thread has devolved into another attack on Nate's math nt

[jefferson_dem]

36. It's discussions like this that separate those who choose to live in the reality-based universe...

from those who operate in a state of sheer denialism.

[DemocratSinceBirth]

40. The Indisputable Truth Is Nate Silver Was Viewed As An Oracle Here In 2008

~

[Blue_In_AK]

41. Polls are voter suppression tools. n/t

[polichick]

45. I agree. nt

[bigwillq]

46. So what does it say about the DUers

that seem to be obsessed with the polls? :shrug:

[Blue_In_AK]

47. Umm, they're nervous Nellies who like to wring their hands a lot?

I say just get out there and vote your conscience and stop worrying so much about what other people are doing.

People here in Alaska are obsessing about polls, too, especially the ones showing Joe Miller ahead by a point or two. Everyone's all "Oh, no, we have to write in Lisa Murkowski so evil Joe doesn't win," as if her votes would be any different than his. I'm for Scott McAdams all the way and to hell with the polls.