« Des Moines Register Poll: Kerry Leads by 3 in Iowa | Main | Analysis of Zogby State Tracking Polls »

Stan Greenberg Speaks!

And when he does, I listen. Here's the full text of a very instructive memo that Greenberg sent around today, with the title "Race stable on final weekend: Review of public and overnight polls":

To give some perspective on the race in its final weekend, we might recall what Wolf Blitzer said on the day before the election, 2000: "And now, let's take a look at the latest poll numbers. The new CNN/USA Gallup Tracking Poll results are being released at this hour. It shows George W. Bush with 48 percent, Al Gore 43 percent, Ralph Nader with 4 percent, Pat Buchanan with 1 percent. And those numbers are similar to other tracking polls," going on to cite the Bush lead in polls for ABC, Washington Post, NBC-Wall Street Journal, CBS and MSNBC-Reuters- Zogby.

In fact, the public polls in 2004 show a remarkably stable and dead-locked race, with Bush stuck somewhere between 47 and 48 percent, short of what an incumbent should have to secure re- election.

The Democracy Corps has a new poll, conducted Friday night and Saturday morning. While the full survey will be completed on Sunday, the half-sample of 500 interviews conducted after the release of the Bin Laden tape, show the race unchanged compared to a survey completed Thursday night. The partial survey shows Kerry at 48 percent and Bush at 47 percent. Like the survey conducted before, it shows the two parties with equal numbers of party identifiers.

The Saturday respondents (250 interviews) were asked the following question: "I'm going to read you a pair of statements about the release of Bin Laden's videotape. Please tell me which one comes closer to your view.

-- It makes me think that George Bush took his eye off the ball in Afghanistan and diverted resources to Iraq.

-- It underscores the importance of George Bush's approach to the war on terrorism.

By 10 points (46 to 36 percent), voters were more likely to think that Bush took his eye off the ball. (These results will be updated when the full survey is completed on Sunday.)

The stability in this poll reflects the overall stability of the race for president. This past week, George Bush polled 47.9 percent as the average of the public polls. That represents only a .5 point change compared to the prior week. Indeed, if one looks at the polls released Saturday and including polling after the release of the Bin Laden tape, Bush’s vote stands at 48 percent in one (Newsweek), 47 percent in one (Fox), and 46 percent in two (Zogby/Reuters and TIPP). That is a weaker result than for the polls released earlier in the week and prior to last weekend.

(These results are based on the results for registered, when available, as that is consistent across polls and has been a better predictor of the final outcome.)

Kerry’s vote was also stable at 46.4 percent on average, up .1 percent compared to the previous week. That is a dead-even race, where the undecided will play the final role, as they almost always break heavily against the incumbent.

The undecided in the race is also stable. During this week, the average of the public polls was 4.4 percent. The polls completed for Saturday have an average undecided of 6 points. There is no evidence of undecided narrowing on this weekend or as result of recent events.

The Democracy Corps combined polls as of Friday showed that the undecided (prior to being pushed to a preference) leaned toward the Democrats by two-to-one and favored a significant change in direction over continuing Bush’s direction by 58 to 29 percent, also two-to-one.

The tracking for the Kerry campaign, conducted for the whole battleground and in key battleground states at the end of the week, including Friday night, show Kerry with a clear and stable lead.

Bottom line, amidst the intensity of campaign’s final days, it is important to keep one’s eye on the stability and structure of this race, with Bush still short of what he needs to win.


I could not be angrier at the fact that the Republican glee at the release of the bin Laden tape is not being answered.

Someone talks about it here.


I always defer to the grownups who have polling data, but I would like an explanation.

This seems ridiculous.

On Sunday, the tracking polls continue to move nicely in Kerry's favor!

Among RVs, FoxNews now has Kerry AHEAD by 2, 47-45 (and tied among LVs 46-46). The Washington Post released their poll early Sunday, and among RVs Kerry is ahead 48-47. Zogby has the race tied among both RVs and LVs.

And once again, Bush is stuck at 47% (at most) when all of the tracking polls Saturday and Sunday are averaged together.

On to victory!!!

They are hoping that people think of Bush like they did on September 12th, when he got his legs back. They're hoping that you also forget everything you learned about September 11th. The lack of action. The confusion. The lack of attention to bin Laden. They hope that if you're thinking about terrorism that you will vote for Bush.

It's risky, but really, at this point what do they have left? They obviously don't have "in a freezer" as so many have assumed.

Since I live in the "reality-based community" I also see this as utterly ridiculous. But again, what do they have left? They have no results to run on...

Did anyone notice the internals on this Dem Corps poll? On the "Who did you vote for in 2000?" question, there was a 6% edge for Bush and yet Kerry still led this poll by 1%. In addition, the poll shows 41% self-identifying as conservatives. I think this is higher than the population as a whole.

This suggests to me that this polls in particular, and pollsters in general, are not getting a truly representative sample, and that Kerry is going to do significantly better than the polls indicate.

The only reason the Republicans crow about the reappearance of Bin Laden is that it might change the subject of the election away from domestic policy issues and Iraq where Bush’s positions are increasingly unpopular, to the issue of terrorism, where he still holds a lead. But that reflects Republican desperation, and could easily backfire. I mean wasn’t this the guy Bush wanted dead or alive? And wasn’t he supposed to be produced by the Bush Administration and not Al Jazeera for an October surprise? It’s hard not to agree with yesterday’s Boston Globe (Bush’s Deadly Blunder) editorial that both the Bin Laden tape and high explosives missing from Al QaQaa point to the Administration’s incompetence in prosecuting the war on terror, i.e., the failure to plan effectively and send enough troops to seize dangerous arms caches, and the diversion of U.S. forces from Afghanistan to Iraq, and the reliance on warlords to catch Osama when he was around Tora Bora. Indeed the Administration’s lack of competence and honesty regarding these matters may have created an Al QaQaa/Al Qaeda “axis of errors” which could conceivably be discussed to criticize Bush’s stewardship of the war on terror by grass-roots activists in the all-important ground game of the next three days.

After all, as Ruy’s post says, the DemCorps poll found about 5% of voters still undecided. From my canvassing and other conversations, I think that number, and especially the percentage of “persuadables,” might be twice that, making the ground game even more important than assumed. Strolling down memory lane, I found some
interesting stats in the 2000 National Election Study. In that election, it appears, if the study and my reading of it are accurate, that 5% of voters made their presidential voting decision on voting day, and another 11 percent made it “in the last few days of the campaign.” Even if these percentages have been cut in half by this election’s starker issue position contrasts and greater polarization, that’s still a lot of people. See you canvassing!

Meet The Press:

Charlie Cook stated that he would NOT be suprised if the
turnout reached 125 - 130 million (this was echoed by
several other analysist).

Charlie Cook also stated that he has been all over the country and has never seen anything like this before.

I think what we see on Tuesday and what we have seen in the tracking polls and opinion polls is going to come down to one crucial difference: GOTV with new registrants.

If the Democratic Party, ACT, Moveon, LCV, and all the other allied groups can GOTV on 80% of the new registrants, Bush is history. I don't buy into the idea that we're going to see close to 100% of new registrants vote (since simply registering to vote when someone brings you all the materials and information you need is not the same as going to early vote or vote on election day).

The wonderful thing is that we've had so many new registrants since 2000 (that most likely voter models ignore in the polls) that if we can turn out around 80% of them for Kerry, we'll be in good shape.

I'm more concerned about turning out new registrants in the Senate races than I am in the race for the White House...

I think the Elephants are really in "wish" mode now. They are hoping that those in the "security first" mode are going to see the Osama tape and switch but that misses some reality.

First, a fair percentage of those "security first" people support Kerry. Bush does not have a 100-0 lead on the "terrorism" issue.

Second, most of those who have this inclination have already made up their minds. People who vote on an overriding single issue (i.e. terrorism, abortion, the environment, etc.) make up their minds early in the campaign.

Those late "undecideds" aren't single issue voters. As a result, character issues have greater sway on them than anything else. This helps explain the late loss of support by Bush in 2000 after his drunk driving charge was made public. There has been no late character issues raised in this campaign so I suspect that we will see a more traditional swing to the challenger in 2004.

On the internals of the Democracy Corps poll, is it possible that either some a-political people don't correctly remember who they voted for in 2000, and assume they must have voted for the winner, or that they don't want to admit having voted for the loser, bad as Bush has been?

On the race itself, it is inconceivable to me that if Gore, a relatively lackluster candidate with a mediocre operation and not much enthusiasm behind him was able to beat the relatively unknown GWB by 500,000 votes, how is it that we can lose this time with a better candidate, an operation that benefitted from the experience of 2000 and a highly motivated electorate and volunteer corps against a known incumbent? It makes no sense that we would lose. Keep doing what we're doing for 2 more days and we have it by 3% (or 3.5 million) and 88 EVs.

If you wanted a clue, in advance of the polling, of who would EXPECT to be more hurt by a resurfacing of bin Laden, there was this quote in the Saturday LA Times. p. A5. It was something Cheney said on Oct. 22:

"We haven't seen much of him. You'll notice there haven't been any Bin Laden tapes running on the air where he's out broadcasting messages, frankly, because we think he is in a deep hole someplace, in hiding."


The last thing they wanted--or, apparently, expected--was for him to appear and remind undecided voters he was still alive and well.

On the Osama tape:
First, it DOES tend to help the Republicans, by focusing on the one issue where polls show Bush has a clear (if reflexive and perverse) majority. However -- two other points. First, many Republicans, as Dana Milibank recently noted in the
Washington Post have "long insinuated and occasionally asserted" that Al Qaeda wants Bush defeated in the election (Krauthammer insisted on this point in a column a few weeks ago in the Washington Post). But now that solid evidence surfaces that Osama seeks to influence the election to help Bush (who has been less vigorous in going after Al Qaeda than Kerry proposes while handing them "Christmas" in response to the 9-11 "Christmas for Tories" in the form of the Iraq War) neither liberals nor the mainstream press dare breathe a word suggesting so. The Kerry campaign, of course, could never and should never so speculate, but liberals are so much more (enforcedly) squeemish than the Right that this is what decides elections.
Second, the notions that Osama either had no interest in influencing the elections, or didn't think this would help Bush are both implausible. Constant spinning to suggest that it WILL help Kerry obscures this fact. If voters realized WHY Osama wants to help Bush, THEN Bush would be in VERY serious trouble, and so would the Repubs in Congress. This whole pattern of discretion (incidentally FOX News apparently suggested, in the teeth of the evidence, that this was trying to help Kerry -- causing a flap on Friday with the Kerry campaign.)

iam watching your comments post election and still you dont get it.you spoke of your daughter commenting that "i'm glad i live in a city that only 9% voted for bush.the only difference between "the city" your daughter refers to and hollywood,is the famous faces.you people live in a bubble,you are so tightly congested,that it is impossible for one to form their own opinion.if you notice,all of the states that voted kerry were tightly congested,even though some of these states are less populated,it is a cold region and the majority live in the cities.thus the onesided views.while the people who live in southern states are generally more spread out.these people,form their own identity,and their own opinions.and these opinions generally line up with the more conservative view of life itself.there is no political diversity in the northeast persay.while in californiathe cities and elitists vote kerry,if you took a poll of the people who live in the eastern mountainous regions of ca.you will find what i say is true. thank you,les davis