More poll madness (sorry, K!)
My Democratic friends are excited by Kerry's narrow electoral majority on electoral-vote.com.
Kerry: 291 Bush: 247
Actually, this map looks very similar to a prediction I had made long ago, with the exception of Florida. I've said it before and I'll say it again, Ohio is more at risk for the Republicans than Florida this year.
The Democrats shouldn't get too carried away, however. Take away Kerry's razor-thin majority in Florida, and Bush wins narrowly.
But it's even worse than that. This map reflects new battleground state results from John Zogby, whose numbers have consistently favored Kerry throughout the campaign. Normally when a fresh batch of Zogby data is added, this map looks like a smurf bled all over it. Today, however, it looks much the same as it did yesterday. Why? Because the most recent polling cycle has given Bush his best Zogby showing yet.
I also have high regard for Fox News polling. No, really! Stop laughing. Fox has actually been nearly as tough on Bush as Zogby has, yet today they're showing Bush with a five-point lead nationwide, as well as in the all-important Ohio.
It ain't over yet, and either candidate could still win this one. Anybody who tells you otherwise is dreaming. But if you force me to bet, my money's still on Bush, as it has been all along.
Comments
Below is a conservative site that shows polls across the country. Note that right now there is a discrepancy between the national poll numbers and the state numbers. Their averages give Ohio to Kerry, but Florida to Bush. The challenge for Kerry, however, would be to win a very close election in at least one of the states, given the fact that both states have Republican governors.
http://www.realclearpolitics.com/bush_vs_kerry_sbys.html
Posted by: PE | October 20, 2004 09:37 AM
Jay Cost's blog (http://jaycost.blogspot.com) is a "must read" for all polling site watchers. I think his analysis is just a tad over-ambitious at times, but he does a good job of illustrating the limitations of polls, polling and pollsters.
He seems to think that virtually every major poll is failing to catch important trends in voter registration this year because they are fixated on their own assumptions from the 2000 election.
As with polling itself, Mr. Cost needs to be taken with a grain of salt (and perhaps a wedge of lime and a shot of tequila) but still well worth reading if you are as obsessive about polling as I am.
BTW, PE, why do you think RCP is conservative?
Posted by: withoutfeathers | October 20, 2004 09:52 AM
Are you joking regarding RCP? Look at the articles they post on the front page. It is obviously a conservative site, just as electoral-vote.com is managed by a Kerry supporter.
I, by the way, work for a major polling company. I am well aware of how polling has become less accurate in recent years, largely due to a declining response rate but also due to using cheaper methods and smaller samples. The problem with averaging polls, in my view, is that it does take into account a possible systematic bias that could be contained in all the polls. Polls, I think, can measure trends, but a systematic bias might mean that the actual number is consistantly off.
Posted by: PE | October 20, 2004 10:07 AM
Correction: does NOT take into account a possible sytematic bias..
Posted by: PE | October 20, 2004 10:09 AM
RCP's commentary is no doubt conservative, but their compilation of poll numbers is straightforward and unbiased. I agree with PE that polls are most useful in illustrating trends, and that's why I pay more attention to the daily tracking polls like Rasmussen and Zogby than any of the others.
Posted by: Barry N. Johnson | October 20, 2004 10:14 AM
No, I'm not joking about RCP. Today's articles might lean right, but scroll down a bit and you will see links to The Guardian, The Boston Globe, Detroit Free Press...GEORGE SOROS!
BTW, I never indicated which side of the fence I thought RCP is on; I just asked why you think it is conservative.
Posted by: withoutfeathers | October 20, 2004 10:21 AM
I have checked out RCP's articles for months now and they've consistantly featured conservatives. That doesn't discredit their poll numbers; it is just an observation. As far as their links down the page, this web site links to Daily Kos and Democrat Underground, albeit under the morons and nuts classification.
Posted by: PE | October 20, 2004 10:33 AM
"[M]orons and nuts classification[????]" I'm going to have to assume we're talking about two differant websites. I can't find that section, or anything resembling that on www.RealClearPolitics.com.
Posted by: withoutfeathers | October 20, 2004 11:07 AM
Another good polling analysis site is http://www.DalyThoughts.com. Gerry Daly is open about his support for BC04, but his methodology is thorough and interesting. He's a good counterpoise to Electoral-Vote.com, and personally, I think he is less prone to cherry-pick polls that support his candidate.
Posted by: withoutfeathers | October 20, 2004 11:17 AM
In the history of the United States, only 4 incumbent presidents seeking re-election have been defeated. A strong 3rd party candidate played a hand in 3, the Great Depression in the 4th. There are no comparable factors this year.
Perhaps we are seeing a new precedent in the making, but history definitely favors a second Bush term.
Posted by: CRB | October 20, 2004 11:17 AM
The "morons and nuts" classification can be found on this web site (cynical nation). I just mentioned that as an example of a site linking an opposing view.
Posted by: PE | October 20, 2004 12:14 PM
I don't think that John Anderson was the determining factor in 1980. Furthermore, LBJ probably would have run in 1968 if not for the Vietnam War. His VP, Humphrey, then lost in a very close election. If anything, the major factor in both cases was due to challenges from within the party.
Posted by: PE | October 20, 2004 12:24 PM
I also think there are more than four incumbant Presidents who have lost.
Jimmy Carter
John Quincy Adams
George Bush
Herbert Hoover
Gerald Ford
Grover Cleveland
Robert Taft
Posted by: PE | October 20, 2004 12:30 PM
PE is correct, my comment referred to the last 4 incumbents to be defeated: Bush, Carter, Hoover and Taft. Ford is the exception, since he wasn't elected in the first place.
Posted by: CRB | October 20, 2004 02:15 PM
Thanks, PE, I was just going through the elections in my head and wondering the same thing.
One small correction: it was William Howard Taft, Bob Taft's father, who lost in 1912. The younger taft, aka "Mr. Republican" lost out to Ike in the 1952 Republican primaries. It turned out to have been a blessing in disguise as he died of cancer in 1953.
Posted by: mal | October 20, 2004 04:44 PM
PE, I have to correct you. While I do indeed have a "Morons and nuts" section, the Daily Kos is not found there. It is in the section reserved for "Lefty bloggers."
Posted by: Barry N. Johnson | October 20, 2004 08:38 PM
The V7 Network - Highly relevant news and opinion for the advanced web professional
Posted by: Internet Marketing | November 24, 2004 07:27 PM