ABC/WaPost Generic Congressional Poll - 53R-40D

[FBaggins]

Oh lordy. This was one of the polls I was waiting for to balance things out.


http://www.washingtonpost.com/wp-srv/special/politics/documents/2010-09-02Trend_Tuesday.pdf

[AndyA]

1. People have very short memories.

Truly, it's like hiring back the drunk to drive your kids around after a near fatal accident because the new nanny hasn't gotten the kids to act like perfect children yet.

So, it's better to have them die in a horrible accident than to give the nanny more than two years to teach them to behave better.

OK...

The Dems need to get together and get their message out. They need to REMIND people WHY we are in this mess. They obviously aren't doing a good job of that so far.

[FBaggins]

Yes. It isn't even "what have you done for me lately"

It's "what do I understand you to have done for me lately - when I only pay attention for short periods of time every couple years"

[golfguru]

26. You may be forgetting the effect of Teabaggers

which were completely missing last go around for GOP.
The energy factor is very high amongst teabaggers and could
explain the generic ballot.

[FBaggins]

28. Were they?

I think that the only real change is the shift from the "christian" coalition (supposedly social conservatives) to pseudo-libertarian fiscal conservatives. Still a minority group driving the larger party.

Sure there's some overlap, but the end result is similar... they can't keep both sides of their party happy at the same time. It may work for this election but it's cr@p for building a governing majority.

[Kdillard]

2. Sure you were.

[FBaggins]

3. They've had a record of bringing in good news when other polls looked troubling.

In May they came in at D+5 right after a string of R+4 to R+10 (including a PPP poll at R+5). In early June they came in at D+3 right after Gallup had R+6 and Rasmussen had R+9. A year ago they were posting double digit Democratic leads while Rasmussen was releasing nonstop polls in the other direction.

Yes... there was reason to hope that ABC would balance some of the polls that seemed too far over the line.

But tell me. I'm interested in this odd psychological phenomenon that you seem to exhibit. Does it make you feel better to attack the messenger? Are polls less valid if you can convince yourself that the poster is secretly gleeful at the bad news?

[bigdarryl]

4. I wouldn't put any stock in any polling it's about BOOTS ON THE GROUND

[impik]

5. People will get what they deserve. This conuntry is so stupid,

[golfguru]

29. Why is the country being stupid when...

Unemployment has stalled at a very high level for 2 years,
The $800 Billion stimulus has not budged the unemployment situation,
50,000 troops remain in Iraq & Afghanistan has double the number troops,
Health Care Reform bill was ugly, with no public option and no restraint on rate hikes,
Yearly budget deficits are now 3 times higher compared to under Bush.

To be fair, the president has made many good regulatory reforms, but
it is the economy, stupid.

[Mass]

6. Looking at the internals, it is clear that the situation is not as clear cut as it seems with these

numbers.

First, it is likely voters numbers. The numbers on the registered voters is a lot closer (D:45, R: 47). (not that it is good, but the gap is big enough to be noticeable).

Second, we are in a situation where a majority of the sample thinks the dems understand them better than the GOP, rate them better on their major issues (economy and healthcare) but apparently, they dont feel they should go and vote.

So, may be it would be time to stop whining and start campaigning. People are not trusting the GOP. They just dont feel they have to go and vote. It may be time to give them a reason to do so.

[FBaggins]

7. I'd say that it's pretty clear cut.

I'd also say that you nailed it on the head. Plenty of people who would otherwise be inclined to vote
for democrats, but for one reason or another have no plans to. The question is how to
get them to change their minds.

[Nicholas D Wolfwood]

8. I still think people are just inclined to keep shuffling the deck til they get a good hand.

Meaning, they're just going to arbitrarily keep changing their votes until life gets better. It's dumb and irrational, but no one ever accused the average voter of being anything else.

[LiberalFighter]

9. I wonder what each campaign is doing?

Are the campaigns out their campaigning like they should be? Do they have their people out doing the necessary work?

[wisteria]

18. Good points. n/t

[Proud Liberal Dem]

10. Without discounting the difficulties we face

How meaningful are generic polls at this point when we know who nearly every candidate is going to be? :shrug:

[FBaggins]

11. In most cases (at this point), they're the best we can get.

Yes, we know who the candidates are in most cases, but a statistically significant sampling in each of 435 House districts (or even just in those 50-100 that are marginally competitive) is a massive undertaking. The non-automated pollsters in particular are just not set up to do this.

What they do know how to do is select a statistically significant nationwide sample and extrapolate what that historically means to individual races.

[Teaser]

13. what we need to do

is look at the probabilistic coupling between individual districts polling, historically, and the national race. Some districts act more like the nation overall, and some are idiosyncratic.

I'd be willing to do the statistical legwork, if someone can point me to a repository of this information.

[FBaggins]

14. No need.

There are professionals who do this for a living.

Nate silver is one of the best, and his House model is due out pretty soon. He's pretty good about releasing his methodology, so you can adjust for whatever factors you think he may be missing.

[Teaser]

16. it's more fun to do it on my own though

If we're going to get crucified, I'd at least like to get some intellectual satisfaction out of it.

Does Nate release his raw data too? or just the processed data and crosstabs?

[FBaggins]

19. Here's his explanation for his gubernatorial model

with links to additional details

http://fivethirtyeight.blogs.nytimes.com/2010/09/05/governor-forecasts-understanding-the-model/

[Teaser]

20. heh, if this goes as badly as my March Madness Regression Model

I could look REALLY stupid.

[FBaggins]

24. Then I hope you predict a republican tidal wave.

:-)

[Teaser]

12. Not really buying any "likely voter models"

this year.

Too weird a year.

So in short, I have no idea what to think.

[FBaggins]

27. My theory on that.

Most (but not all) mid-term years, republicans tend to outperform their generic polling numbers by a couple percent. It's hard to identify how much of this under-performance in the polls is accounted for by registered/likely voter models and how much is based on their greater propensity to turn out - because both of these tend to show up at about the same time in the majority of polls.

My theory is that the stronger enthusiasm they've been running this year has caused that second factor to show up early this year. It also seems that more pollsters (in response to Rasmussen, Survey USA, and PPP) have shifted to LV models earlier in the cycle. What that could mean is that while the polls are "accurate" - they are the equivalent of a 3-5 point smaller lead when compared to prior years' polling.

Still a problem, but not nearly as large a problem as it appears when compared to prior cycles.

[jefferson_dem]

15. Batten down the hatches, friends...

There's a cold wind blowing ... and it's not going to be pretty for us (the good guys).

I'm hoping we can stem the rising tide just a bit.

[wisteria]

17. This is the news media patting themselves on the back for a job well done.

They have been successful in swaying the public opinion in favor of Repubs. And, the people being swayed are ignorant of facts and aren't use to having to wait for anything. If Repubs do manage to gain significant ground, they are going to have to actually do something positive for the economy-which we all know they are incapable of doing.

[Aramchek]

21. sure you were. you aren't fooling anybody, frodo.

[KingFlorez]

22. Right on

[thewiseguy]

23. Yea, lets shoot the messenger!

:eyes:

The cycle is so predictable. Poll posted with favorable results, the thread gets lots of Recs and everyone is happy.

Poll posted with unfavorable results, the polling methods are questioned, the messenger is accused of many things, and the news media are to blame!

[FBaggins]

25. See #3

What a crock.

[golfguru]

30. naw...let's just bury our heads under sand

then we won't need to see what's going on in political situation.