« Get Out the Vote in Pittsburgh - It's Your Turn! | Main | Report from Obama Rally in Springfield, IL »

Monday, November 01, 2004

General Election Cattle Call, November 1: Final Projection

Posted by Chris Bowers

National Popular Vote
John Kerry: 50.13%
George Bush: 48.37%

Others: 1.50%

Interestingly, it is the TIPP poll that gives Kerry the final push over 50%. With 9% undecided, and most of the undecideds allocated to Kerry, Kerry���s lead rises from 0.40% without TIPP to 1.76% with TIPP. In front of many thousands of people, I am obviously staking my reputation as an election forecaster on the Incumbent Rule. I stand by my research on the subject, and I expect it to bring Kerry into the White House with a very 1976-esque margin of victory.

Electoral College
John Kerry: 291
George Bush: 247

States changing hands from 2000: FL and NH switching to Kerry

States Projected By Less Than Three Points

Ohio: Bush 50.2, Kerry 49.4
Wisconsin: Kerry 49.7, Bush 48.4
Florida: Kerry 50.0, Bush 48.5
New Mexico: Kerry 50.3, Bush 48.5
Iowa: Kerry 50.7, Bush 48.5
New Hampshire: Kerry 50.6, Bush 47.8

According to my methodology, every other state is projected by more than three points, and thus leans toward one candidate or the other. Every single other state goes as you would expect. I have Kerry winning all 291 of his electoral votes by more than 1%. In other words, I project Kerry as the winner tomorrow night, no overtime. I have Kerry falling short in Ohio but not by much. Please, prove me wrong in Ohio!

That is the best guess I can make. Even though I allocated 86% of the undecideds to Kerry, I think I erred on the conservative side when it came to which polls I included. I always used LV models when it came to state polls, and I included, without any adjustments, polls such as Mason-Dixon, Gallup and Quinnipiac that I felt over-sampled Republicans.

If I am nervous tomorrow and have nothing else to do, I will adjust this project to take into account the final national polls. However, I doubt it will have much, if any difference. I think Kerry is going to win. I think there is going to be record turnout. I think the undecideds will break overwhelmingly for the challenger. I think our GOTV efforts will make an important difference in Kerry���s favor. We will all find our tomorrow how close I came.

Posted at 07:55 PM in General Election Cattle Call | Technorati

Comments

Thanks for all your hard work in doing these projections, Chris. It will be amazing once we finally see how it all turns out.

Posted by: DavidNYC at November 1, 2004 08:18 PM | Permalink | Edit Comment | Delete Comment

Since 1984 undecideds have gone 34% to challenger 32% to incumbent, 12% other and 22% don't vote.

Posted by: Pete at November 1, 2004 08:22 PM | Permalink | Edit Comment | Delete Comment

Thanks very much, David and Chris, for all of your hard work on this site. Your good information, analysis, and optimism have helped to sustain my fragile hopes during this crucial election. The first presidential election I voted in was twenty years ago, and it was a bitter disappointment. I remember thinking that it would be terrible and dire for the country to have another 4 years of Reagan. It wasn't as bad as I feared, but I think with Bush it would in fact be terrible.

I still would like more information as to why you are allocating so many of the undecideds to Kerry. Andrew Kohut (sp?) was on PBS tonight, and he noted that Gallup allocated 90% of them to Kerry, but that he didn't for his poll (which I think shows Bush winning). Kohut said it was too close to call, and he could see the election going either way. When I look at this site it raises my hopes, but www.race2004.net shows Bush winning. I'm something of a nervous wreck because I think so much is as stake in this election.

Anyway, thanks again. Here's hoping that we'll be celebrating tomorrow night around midnight EST.

Best, Ben

Posted by: Ben at November 1, 2004 09:24 PM | Permalink | Edit Comment | Delete Comment

Though I haven't been following the polls as closely as most, including yourself, I've been checking the website http://www.electoral-vote.com. It has gone back and forth almost on a daily basis as to who is in the lead (based on polls). They predict Kerry winning by over 300 EVs.

What's your opinion on this site?

Posted by: Mathew at November 1, 2004 09:38 PM | Permalink | Edit Comment | Delete Comment

Chris, I think Ohio will prove you wrong. I get the feeling that today's tracking polls have produced a sea of bad samples. Is this just a gut feeling? Perhaps, but when the Republican pollster Strategic Vision has the best numbers for Kerry out of the lot, I get the feeling Bush's standing in OH is being overstated in the tracking polls. If Strategic Vision has Bush ahead by only two points in Ohio, it's hard to imagine him not winning it.

Posted by: Mark at November 1, 2004 10:07 PM | Permalink | Edit Comment | Delete Comment

Hard to imagine KERRY not winning Ohio I meant to say.

Posted by: Mark at November 1, 2004 10:10 PM | Permalink | Edit Comment | Delete Comment

I've looked at polls all over the net, and, not surprisingly, they have something for everyone! Again, the waters are more muddied than cleared by all these polls. We will finally find out in the only one that counts--tomorrow. I don't have a clue, nor do I have a gut feeling after looking at literally scores of polls this evening. Bottom line: polling is far from a science. I guess that's why the swing states are called swing states--more than polls, we definitely have to take a leap of faith to believe they will vote the way we hope they will.

Posted by: pepe at November 1, 2004 10:12 PM | Permalink | Edit Comment | Delete Comment

I think Chris's analysis is spot on except that it might be just a little conservative: he may be under-representing what I expect to be a huge Kerry turnout advantage (pretty much suppressed by the likely voter models in many polls). Check out Ruy Tiexera on the racial and age bias inherent in the Gallup likely voter model at:
http://www.emergingdemocraticmajorityweblog.com/donkeyrising/archives/000808.html

Anyway, our picks of the Electoral College are identical except that I give Kerry Ohio (there, Chris!). So I end up with a 311 EV count for Kerry.

If you want to make your own electoral college prediction (and maybe win a nifty board game) at the Race 2004 site. The vast majority of predictors (about 3-1) foresee a Kerry victory of varying proportions.
http://www.race2004.net/contestcover.html

Anyway, I recommend trying your hand at entering. It's fun and easy.

Posted by: TomNY at November 1, 2004 10:12 PM | Permalink | Edit Comment | Delete Comment

I agree with TomNY. I have a feeling that the polls are underestimating Kerry's support (due to undersampling of young voters and assumptions about "likely" voters).

If I'm right, and this is a nationwide phenomenon, then it could be an easy victory for Kerry, in which the only states that remain uncertain into Wednesday will be states like Colorado, Arkansas, Missouri and Nevada (as opposed to FL and OH, which would be close but for Kerry).

If I'm wrong and the polls and pundits are correct, then it really is too close to call, but I still think Kerry should be able to pull it out with Florida OR Ohio.

Keep this in mind: only once has an incumbent (even an incumbent party) won a close election in the last century. [What I consider close elections are 2000 (Gore lost...sort of), 1976 (Ford lost), 1968 (Humphrey lost), 1960 (Nixon lost). You've got to go back to 1916, when Woodrow Wilson narrowly won re-election.

Posted by: Jason at November 1, 2004 10:46 PM | Permalink | Edit Comment | Delete Comment

Zogby's prediction in Election 2000 part 2:

Kerry 264 (NH goes to Kerry)
Bush 247
Tied FL

Posted by: DFuller at November 1, 2004 11:03 PM | Permalink | Edit Comment | Delete Comment

Oh, one more thing . . . if I'm right, then Kerry could bring a few senators in on his "coattails", most notably in NC, SC, FL and especially CO.

Posted by: Jason at November 1, 2004 11:25 PM | Permalink | Edit Comment | Delete Comment

Jason, at least CO is more of an "inverse coattails" effect. Ken Salazar is more popular here than John Kerry.

Posted by: Ilana at November 1, 2004 11:33 PM | Permalink | Edit Comment | Delete Comment

Ilana, do you suspect Salazar will win in CO? I was sweating for him last week when it appeared as though Coors was gaining steam. Now it looks as though Salazar's back on track. What say you?

Posted by: Mark at November 1, 2004 11:48 PM | Permalink | Edit Comment | Delete Comment