Gallup applies their likely voter screen and it's a massive adjustment (R+13 to +18%).

[FBaggins]

Actually very little movement among Registered Voters.



PRINCETON, NJ -- Gallup's generic ballot for Congress among registered voters currently shows Republicans with 46% of the vote and Democrats with 43%, similar to the 46% to 46% tie reported a week ago. However, in Gallup's first estimates among likely voters, based on polling from Sept. 23-Oct. 3, Republicans have a double-digit advantage under two separate turnout scenarios.




These initial estimates are based on interviews with more than 3,000 national adults, including more than 2,700 registered voters, and more than 1,800 adults who demonstrate a high probability of voting this fall, based on their answers to Gallup's standard likely voter questions that probe current voting intentions and past voting behavior.

[craigmatic]

1. I don't buy the likely voter numbers because they only count people who voted the last few cycles

but the majority of Obama's voters just started turning out in 2008. They can't predict voter behavior when they're not neccessarily counting them.

[FBaggins]

3. Historically, that has proven to be an accurate measure.

If anything (in midterms), it has been to our advantage.

but the majority of Obama's voters just started turning out in 2008

Nothing even close to the majority, but your point is still valid. Many Obama voters were first time voters.

It is, however, important to note that even their tightest LV screen still results in far more people counted as likely voters than actually turn up to vote. In this poll, it was 2/3rds of the registered voters polled... and midterms have been around 40% turnout in recent years.

[WeekendWarrior]

2. More sky is falling news

Nothing to see here.

[FBaggins]

4. Indeed. It's likely that little has changed.

This one is almost certainly way too high, but we've had a few that looked more favorable in recent days. It's likely that the real picture is somewhere in between with nowhere near as much movement as some would like/fear.

One thing that is clear, however, is that turnout can make a several-point difference in this race.

[MannyGoldstein]

5. Nothing to see except the falling sky

DLC=Democrats Lose Congress. It's been true in all six elections when they've controlled the Democratic Party.

[Aramchek]

8. grow up, mannfred

[CoffeeCat]

6. I think that makes sense...

...because there are many Democrats who are registered--who may just sit this one out. There are many
more Democrats who are registered and may have even been active in campaigns--who feel unhappy or
not as enthusiastic about voting.

The stats make perfect sense to me.

I have volunteered for the Democratic party during the last several election cycles. I was an absentee
ballot courier. In 2010, I was an Obama precinct captain who was heavily involved in the campaign. I wrote
a speech, gave it at our caucus and worked aggressively to turn many to Obama's side. My work at the caucus was
featured in a short article in Newsweek. I also went door-to-door campaigning and I made phone calls.

I have no desire to vote in November. For the first time since I've been 18 and eligible to vote, I may
miss an election. I just feel that it doesn't matter any more.

I don't like feeling that way, but those are my genuine feelings.

[MannyGoldstein]

7. I feel your pain, but...

We voted for FDR, we got Hoover. That sucks.

But even Hoover and company were far better than Hitler.

Get out and vote for the folks who suck less, and let's primary 'em next time and get real Democrats.

[prolesunited]

31. You're comparing Obama to Hoover?

Who was your pick on the primary ballot?

[MannyGoldstein]

34. Yes. For more details, see...

http://www.harpers.org/archive/2009/07/0082562

I voted for Obama in the primary, and contributed to his campaign.

[dionysus]

32. bless your little heart for trying so hard.

[Drunken Irishman]

10. Your mindset is what's wrong with this country...

And it's unfortunate your selfish attitude will fuck us all.

[depakid]

11. Rather than blame and berate an individual, it's generally more useful to grasp why

people (and groups of people) hold or express these sorts of feelings and views- and what sorts of things might be helpful in motivating them to support your party or your cause(s).

[qazplm]

13. we already know

but it's likely to be something that ignores the reality of the situation, which is that bad democrats are light years better than republicans, and that at this critical point in our history, we simply cannot allow republicans to come back in charge.

[cliffordu]

14. What my mildly inebriated friend says.

+1000

[HughMoran]

15. Yay, another downer post before the election

:eyes:

[MineralMan]

25. If you don't vote, you're irrelevant.

If you encourage others not to vote...well...

[prolesunited]

36. And Republicans in charge will be better how?

[depakid]

9. A good question when it comes time for post mortems will be "how accurate are these surveys

in predicting likely voters- or demographic or partisan patterns of likely voters?"

[Writer]

12. I think this is a very, very serious issue.

I wouldn't take these numbers for granted. Obviously Democrats don't plan to vote. Obama needs to continue campaigning across the country. He needs to keep doing interviews and reminding the American people what he has accomplished. Hopefully a few naysayers will come out on Nov. 2nd.

[wiggs]

16. OK. Good to know...so what's DNCs strategy for overcoming this disadvantage? It's

not like dems can't adjust, if they want to, or change strategy, or start buying airtime, or get tougher about the horrendous gop record in and out of power, or get high profile supporters, or get more gotv efforts, etc..

We're down and have to try harder. Let's assume the bad news is correct and pull out all the stops.

[Mass]

17. The only good reason to consider these numbers is if they make us fight harder.

There are serious issues with all these polls, RV or LV, some of which are exposed by Nate Silver today. But I welcome these numbers if they can wake up some of the idiots who think they can afford to stay home.

[Atman]

18. It's bullshit. Gallup admits to skewing towards GOP because it believe there are more GOP voters.

It's insane that anyone listens to them. They don't RANDOMLY sample people. They look for specific demographics, because Gallup believes more Americans are righties than lefties. So they generally try to sample 5% more repubs than dems, because they believe that is a more accurate representation of America. It is PURE BULLSHIT. Gallup works for the GOP, even if they themselves don't know it.

.

[Robbins]

19. Gallup

Time to officially Ignore gallup.Even RAS doesn't have lead this big.And Likely voters are flawered.The One Nation rally was
bigger than Beck despite what RW Pundits try to claim like on Ed today.

[depakid]

20. If people are going to follow that advice, then they have to do it across the board

And that means even when the purported results say something that people want to hear or believe.

Personally, I don't think most folks are capable of that sort of restraint and consistency.

[DemocratSinceBirth]

22. +1

Cherry picking polls is for amateurs. But obviously a double digit deficit would have catastrophic consequences for Democrats.

[depakid]

26. What Nate's been trying to do is raise awareness about the way polls are done

This poll, for example is nearly meaningless on its own- and the reason why that is comes down to methodology.

The bogus bits are a generic ballot, using multi-stage clustering and purporting to represent (through a few sample locations, weightings, etc.- which we don't know much if anything about) and applying it broadly (asserting external validity) to reflect every states' races- or the likely sum of all of those races.

There are so many reasons why this type of analysis doesn't say what people and media corporations assume that it does that it would take 12 pages to summarize- and we really should applaud people like Nate who have the gumption and the ability to make the attempt- and make it accessible.

That said, the poll's not useless- especially over time. It can for example reflect fairly broad trends.

[golfguru]

21. There are 35 million adults without a job

It must be difficult to get energized enough to go vote
for them.

[book_worm]

23. If the GOP did win by 18-points they would gain something like 100 house seats and who knows how

many senate seats. This likey voter scenerio has been shit and it's shit being fed by corporate pollsters in an attempt to keep dems unenthused and to question why they should vote if we're going to lose anyway.

[Jennicut]

24. I just think it is highly unlikely.

And very hard to predict voter turnout in general.

[Atman]

27. Actually, no. It isn't that difficult. We used to have real polling.

Even real exit polling. With the introduction of BushCo and mysto-electronic voting, all the media stopped doing Exit Polling. Remember? Are our memories that short? We used to be able to nail this stuff down within a point or two. Now, with the introduction of electronic voting, everything is a mystery, every race is 50/50 right up until election night. It's pure, 100% BULLSHIT. The same people that own the media own the polling companies and own the voting companies (yes, we have private corporations that control our votes now, instead of local election boards). Deal. It's a recipe for revolution. The righties want it, they've planned for it, they expect it. That's why they've socked away the entire U.S. Treasury in their private bank accounts...hedge funds, betting against America.

.

[Bluenorthwest]

29. Sorry, polls never had nor could have the accuracy you claim

to remember. A point or two? Nope. And my local election board control all of our ballots and election matters. So don't 'we' that which is a local issue. Your votes might be in private hands, not mine.

[Atman]

33. If your local election board uses an electronic machine...

...you DON'T have control. Someone can tap into those machines and change votes. You couldn't do that with lever machines. I know, I was an Assistant Registrar of Voters in my town. My last year doing it was the year they went electronic, by order of the Secretary of State. And yes, exit polling is extremely accurate.

.

[Arkana]

28. The problem here is it's always "Generic D" vs "Generic R"

when this is almost never the case. Every election has peculiarities unique to it and Gallup isn't taking ANY of those into account.

[Bluenorthwest]

30. When we know the candidates, the only reason to do 'generic'

questions is to skew the results. Period. If a fantasy candidate was running against another fantasy candidate in an election today, not the real one, who would win? If, if, if. If today was the election, and Angle was really Ronald Reagan, would you vote for Ronnie or some random Democrat? That is the question as heard by Republicans. But the election is not today, and the candidates are not 'an Republican' vs 'a Democrat' at all. What reason could there be to ask a generic? The actual names are there to poll. Why generic? For to get the answer they want.

[depakid]

38. The reason they're doing generic here is to get a result that covers nationwide sentiment

if you want to see similar results using amalgamated state by state poll results, look at Nate's site.

[Imajika]

35. These numbers are absurd..

Gallup's Likely Voter screen here appears utterly broken.

I agree that the right is very motivated, and turnout may be somewhat of a problem for us, but this Likely Voter result is not in any way believable.

Sorry Gallup, try again.

[MadMaddie]

37. Yawn....

They can do all the polls they want.

This election will be based on Voter turnout!

Get out and Vote in November!