Posted on 10/20/2008 5:13:17 PM PDT by WatchYourself
Don't be discouraged by polls showing McCain and Palin trailing Obama in key states. In 2004, one state poll after the next had underestimated Bush's support, often going well beyond the "scientific" margin of error. Meanwhile, those same polls either correctly predicted or over-predicted Kerry's numbers. Out of over 100 state polls reviewed, more than 90% of polls undersampled Bush, while more than 65% oversampled Kerry. Some finding from key states:
Florida
In 2004 Bush won by 5 Points (52-47)
Of the 9 major polls taken right before the election, Three had Kerry winning by 2-5 points, Two had a tie, and the other Four predicted a Bush win, but underestimated Bush's strength by an average 2.25%. Overall, Bush was underestimated by almost 5%. Most of the polls Margin's of Error were between 3-4 points, meaning that Bush outperformed even the maximum MOE estimates. Rasmussen was closest, correctly predicting Kerry's 47%, but underestimated Bush by 2%.
Missouri
Bush wins by 7% (53-46%).While Kerry's polling figure was accurate (an average of 45.3), Bush was again under-estimated by 3.5 points (polled average of 49.5). The final SurveyUSA poll was most accurate (52-47%). Bush exceeded the MOE in about half of polls.
Georgia
Bush wins Georgia big (58-41%) . There were 4 polls taken during the final days of the election. In a continuing trend, all four correctly predicted or over--predicted Kerry (Over-predicted 1%) while Bush polled below his figure (by 3-6%). 2 of the final 4 polls had Bush exceeding the Margin of Error once again.
Indiana
Bush wins by 21% (60-39%). The two polls taken closest to the election once again correctly predicted or over-predicted Kerry (SurveyUSA was spot-on at 39% while Research 2000 over-predicted by just 1%. But the news, once again, is that Bush was underestimated by 2% and 4% respectively.
In Ohio,
Bush won with 51% of the vote. FOX News and Rasmussen were closest, and just 1% below Bush's actual target. The final CNN/Gallup Poll had Kerry winning by 4 points (50-46%). Despite this being a complete near-flip, Gallup was still within the "Margin of Error" when maxed out in favor of Bush.
The trend continues:
Colorado: Bush wins by 5%. Kerry correctly predicted by Zogby and SurveyUSA (47%), but underpolled Bush by 2 and 3%.
Minnesota: Kerry wins by 3. Zogby correctly gets Kerry's figure but underestimates Bush by 3 while CNN overestimates Kerry by 1 and underestimates Bush by 4%, outside of their MOE for Bush.
Iowa: Bush wins by 1%. The last two polls taken, SURVEYUSA and Zogby both had Kerry winning by 3% and 5%. Bush was underestimated by 2-5% in the final 7 polls taken by various groups.
Virginia: Bush wins by 8% (54-46). The final poll taken in the state was done by Survey USA, and it claimed Bush 51-47%, overestimating Kerry by 1 while underestimating Bush by 3. Rasmussen had Bush underestimate by 4% in it final poll, while underestimating Kerry by just 2 (50-44). Like Rasmussen, Mason Dixon was also error-full in the Margin of Error count, underestimating Bush by 6 (and Kerry by 3).
Hmmmm.... seems to be a pattern shaping up.
Wouldn’t get too carried away with that.. There are undecideds in the polls, not in the votes.
Good work.
“Wouldn’t get too carried away with that.. There are undecideds in the polls, not in the votes.”
I realize this, which is why Kerry’s figures are important. Even with the undecided’s in the polls, Kerry EXCEEDED what he actually got. So either all of the “undecideds” in the polls eventually went for Bush, or the polls had a distinctive bias towards Kerry and against Bush.
Please be careful with this sort of reckless post as many on this site have become addicted to bad news (sorry for the sarcasm).
It seems we disbelieve the media on everything except the polls. Fact is the polls and the media are the same thing. They are trying desperately to keep us home on election day!
A valid and good point, Watch! We know that these daily polls are junk with little or no meaning. They consume more and more news space to cover the fact that more “journalists” have given up the REAL work of news reporting.
This is why I generally ignore all of these polls - they have long since been known to be junk and filler. The REAL poll (and the only one that counts) is on Novemebr 4th. The rest of them are little more than fish wrap.
Jim Geraghty in his Campaign Spot blog at National Review Online has great insight from his sage “Obi Wan” on this point. He called it the Chuck Schumer - Rahm Emmanuel strategy to depress Republican turnout. The have the media foster poll results by oversampling Democrats, combined with the media then hysterically creating the myth that the concocted “juggernaut” is real. It’s why, as Ann Coulter pointed out in her latest column, the media has been oversampling Democrats (and far overestimating their public support) in their polling since the ‘76 Presidential Election.
I’m sure a lot of this has to do with money too. The Drive-bys and the Democrats know all too well huge business interests do not donate to Democrats, logically thinking that Democrat tax raisers are not in the best interests of a business, so they won’t support those that want to take their profits. But the little the Democrats DO get from corporations would be much less if REAL polling data made the headlines - I mean, who’d wanna give money to a lost cause? Why is it that we keep hearing about Obambi’s alleged “$150 million” war chest? If he’s got this in the bag like the drive-bys and the dems want us to believe, and if this “juggernaut” BS issomething we all need to realize as true, then why is it that the Obambi campaign keeps asking for such huge amounts of money?
When I post here that GOP historically outperforms polls, I almost always get one or two negative replies. Its a fact, GOP outperforms the polls, or polls tend to understate GOP support.
Ann Coulter has a great column on this subject (linked at Drudge) where she did some impressive research also. One of the funnier things she said in that column was that while reading the news accounts prior to the 1984 election (Reagan running for re-election) she found herself feeling worried that Reagan was going to lose. The slanted press has been hard at work for decades trying to shape elections. For the younger folks: Reagan won in 1984 by a landslide.
Pollsters tried to cover up their ineptness and/or partisan intent, by claiming that many voters didn't make up their minds in 1980 until they got into the voting booth. How many people do you know who would go stand in line to vote, even if they were still undecided?
Excellent post.
The irony is that all this oversampling is creating what looks to be an Obama easy win, which could all people - namely college students - to get cocky and not show up at the polls.
He’s still running ads in Illinoyd & Missconsin.
“Mispolled”
That sounds like a good name for a movie. These polling shenanigans should be documented in a Michael Moore type documentary - only honest.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.