« Kerry, Bush Tied in New Economist Poll | Main | Harris Poll: Statistical Tie on Eve of Debate »

LA Times Poll: Bush Ahead by 4 As Debates Begin

Bush leads Kerry 49-45 percent of nation-wide RV's, with 6 percent undecided, according to a Los Angeles Times Poll conducted 9/25-28.

Comments

Here's what I don't understand about your commentary: You consistently view ALL polls in scenarios favorable to Kerry. To wit:

You use RVs as your benchmark, implicitly assuming that everyone who is registered will vote, giving Dems the advantage in almost all RV versus LV comparison. Perhaps most important, you invariably leave Nader out out your comparisons, ignoring the fact that Nader is on the ballot in 32 states, including the important battleground states of FLA, PA, NM & WIS.

When Nader is included in the LA Times poll, GWB leads Kerry 51-45.

I also see little or no comment on President Bush's job approval ratings. Excluding Zogby subscriber polls, GWB job approval is in the 52%-56% range, over the critical 50% threshold for incumbents.

A final note: 4 months ago when the Pro-Kerry LA Times had Kerry leading GWB by 8-9 points, you justified Kerry's lead and the overwhelming Dem party ID advantage (6%) in that poll as a true microcosm of the electorate. Now you constantly spin current polls with Rep party ID advantages as frauds, ignoring Gov Arnold's call to Rep arms during the Repub convention and that the war and terror impact on tilting the country more Repub.

I don't get it!!!

Nader is not on the ballot in Wisconsin at this moment. The Wisconsin Supreme Court is hearing arguments this morning.

http://www.channel3000.com/news/3766455/detail.html

He's trying to balance out the might right-wing noise machine, of which you are but a tiny cog.

Hey All,

I believe we all read the polls the way we need to read them to find comfort. That is true for Republicans and Democrats. I read Rasmussen with a clear understanding of the potential bias.

Early on in the process I decided I would rely on three/four polls and stick with them - through the ups and downs. I also read the editorial pages of several papers everyday (split between liberal and conservative - north/south -east/west). I try for balance. In so doing, there are good days and bad days. Overall, the presidential race remains very close.

The areas that that are most hearening are voter registration and "get out the vote" efforts. It is those two areas that I believe will pay the biggest dividends and do not necessarily factor into the current polling trends.

Efforts by Republicans to prevent voters from registering are in line with expectations. One only needs to harken back to Richard Nixon and his "dirty tricks" to understand the mentality of some partisans.

Also of note is MSNBC dismissing Frank Lutz before he has a chance to do a post debate spin with *undecideds* tonight.

After all my reflections I do have a question - how come REALCLEAR is so selective in the Polls used to compute their data?

Jody

Regarding the debates, TNR's Josh Benson offers some excellent advice ( http://www.tnr.com/doc.mhtml?i=express&s=benson093004 ). If there is an opening, Kerry should bend or even break the rules by asking "Shrub" direct, uncomfortable questions! As you know, there will be no rebuttals or follow-up questions because the Republicans are afraid their unraveling Iraq policy in particular would be challenged. But this doesn't prevent Kerry from at least asking rhetorical questions that put the President in an uncomfortable light. Benson offers a few examples:

"Mr. President, Iraq is on the verge of civil war. Entire towns are under the control of terrorists. A thousand American soldiers are dead. Yet you say peace and freedom are on the march. How do you explain this?"

If either Bush or the moderator challenges Kerry for breaking the rules, a handy line would turn the tables right back: "This isn't about rules," Kerry could say. "It's about the right of our soldiers' families to have answers." "If Bush can't handle a simple question, how can he handle Al Qaeda?"

Benson:
"With some tact, this could be a major theme of the evening: Bush's woeful and continued refusal to confront reality in the Middle East. It certainly fits with other aspects of his presidency: the lack of press conferences; his initial refusal to meet with the 9/11 Commission, and then only with Dick Cheney in tow; ignoring foreign leaders who don't say what he wants to hear. Plainly, this is a president hiding from the world. It's about time Kerry called him on it--rules be damned."

The LA Times poll would show a tie if weighted to the party composition in the 2000 exit polls. The poll's internals are at:

http://www.latimes.com/media/acrobat/2004-09/14455239.pdf

Bush gets 7% of Democratic support (multiplied by the D weight of .39 = 2.73), 94% of Republican support (multiplied by the R weight of .35 = 32.9), and 50% of Independents (multiplied by the I weight of .26 = 13), for a grand total of 48.63.

Kerry gets 90% of D support (x .39 = 35.1), 4% of R support (x .35 = 1.4), and 45% of I support (x .26 = 11.7), for a grand total of 48.20.

This jibes well with Rasmussen's latest reading, Bush up 48-46.

"Smooth Jazz" makes a valid point that Ruy may have shown some tendency in the past to be more accepting of pro-Democratic discrepancies from 2000 turnout figures than of pro-Republican discrepancies. However, that has no bearing whatsoever on the question of whether 2004 turnout will reproduce 2000 and 1996.

This election is going to come down to the turnout, folks.

Jody,

You make some excellent points. As far as I know, RCP only leaves out Zogby Interactive; They do in fact include Zogby's phone poll. Their view on Zogby Interactive is that insofar as Zogby subscribers tend to be overwhelmingly Dem, RCP believes Zogby's subscriber poll is biased towards Kerry. I believe they leave out the UK based Economist poll for the same rationale.

They also calculate their averages based on LVs only, even though they do have WITH and WITHOUT Nader numbers. The one potential issue is that they do not include RVs in their averages, ostensibly because the number of permutations & combinations are so many, most visitors to their site would be overwhelmed. They do include the RVs polls in the links on their site; They just do not include it in their RCP averages.

At the same time, they do include all other close polls cited her recently: ARG, IBD, Rasmussen & Harris. They just use LV figures.

http://www.realclearpolitics.com/index.html

Smooth Jazz:

Some comments on your post just MHO

You use RVs as your benchmark, implicitly assuming that everyone who is registered will vote, giving Dems the advantage in almost all RV versus LV comparison.
Reason I believe that is done is because there is little empirical evidence to support LV use this far from an election. The idea of LV came to try to determine vote shortly before and election. Analysis of LV voting in the 2000 election indicates that its use is good only shortly before an election not a bit over a month earlier.
Perhaps most important, you invariably leave Nader out your comparisons, ignoring the fact that Nader is on the ballot in 32 states, including the important battleground states of FLA, PA, NM & WIS.
Do you really believe Nader is an issue in this election? Even in 2000 Nader got 2.74% of the vote. Do you really think that he is going to get more than that this year? The current 3 ways have Nader with at least 2% and in some polls (e.g. Time) at 5 or 6%. Do you honestly think that is going to happen this year? All of the Nader voters I know (~10) from 2000 have no intention of voting for him again. This election is going to be about Bush-Kerry period, Nader is going to have minimal impact, and I suspect he will end up with less than 1%. That said even with Nader Bush is up only by 5% at most
I also see little or no comment on President Bush's job approval ratings. Excluding Zogby subscriber polls, GWB job approval is in the 52%-56% range, over the critical 50% threshold for incumbents.
Yet he still fails to get 50% of vote consistently. This issue is what is he polling. I donít really care if approval is 99%, if people are not voting for him more than opponent it is
A final note: 4 months ago when the Pro-Kerry LA Times had Kerry leading GWB by 8-9 points, you justified Kerry's lead and the overwhelming Dem party ID advantage (6%) in that poll as a true microcosm of the electorate. Now you constantly spin current polls with Rep party ID advantages as frauds, ignoring Gov Arnold's call to Rep arms during the Repub convention and that the war and terror impact on tilting the country more Repub.
The ID advantage is ~4 points now so 2% difference is not that far off. That is a big difference from 7-12% we see in some polls
SJ, obviously this is a democratic site and there is going to be some bias and spin. However, the level of these I see is quite small. The critical service this blog is providing is that the election is still very close, by no means a landside. Republicans have hit the spin that Kerry is on the ropes in his last gasp and only a miracle can save him now. The mainstream press has eaten it hook line and sinker. This has led to discouragement on the side of Kerry supporters. When my wife saw the Gallup poll that had Bush up by 13% she thought Ė on my God this is Gallup, it must be accurate and Kerry is finished! (Eventhouth a 13% lead at this point is not insurmountable). When I showed her the arguments and issues and involved and the point that Bush is at best 4-5% ahead and possible as little as 2% her mood changed considerably.

Ramdan,

That was a well thought out rebuttal to my post. I will not counter rebuttal point by point; however, I do see a trend in the polls:

Almost all of the major, mainstream polls - WP/ABC, Gallup, AP, NY Times, Pew, LA Times, CBS, among others - show GWB polling at or above 50% and beating Kerry by margins in the 6-8 point range; while other polls including Economist, ARG & Harris show the race to be much closer.

Though I've see many reasons (ie party IDs mix, cell phones not covered, etc.) why this disparity exists, I honestly do not know why erstwhile reputable polls differ so much. I refuse to believe however that entities such as CBS, WP/ABC, NY Times, AP & Pew which have historially leaned Dem, to be all in the tank for GWB - And will jimmy their polls in a way that's favorable to him; I just don't buy that.

All I can say is that the race will be close, and, as many here have rightfully said, it will come down to turnout - Provided there is no blatant fraud by either side.

Best Regards,
SJ

Does anyone know *why* Gallup, and now even the LA Times, are weighting toward Republicans? There is no statistical methodology that I know of that can justify it, since the new weightings don't reflect recent history. Do they say they have some reason to expect more Republicans than Democrats to turn out this election? If so, why?