99.94%! Kerry's win probability (332 EV) if he gets 75% of the undecided

[TruthIsAll]

Here is a range of outcomes. Take your pick.

Und% Prob EV
50 94.5 306
55 97.1 312
60 98.5 317
67 99.6 324
75 99.9 332


Blowout!
Tsunami!
Landslide!
Mandate!

Truth!

http://www.geocities.com/electionmodel/

[Dawgs]

1. I doubt he gets 75% of undecideds...

I think more like 65%.

[requiem99]

2. This assumes how much Democratic and Republican turnout?

Because Dem turnout is going to be the highest its been in decades, possibly ever.

[TruthIsAll]

7. There are no turnout assumptions-in the model, but you are right.

Heavy turnout means bigger Kerry numbers.

tia

[requiem99]

9. Uh

There has to be SOME turnout assumptions, how can you predict the result without some sort of turnout assumption factored in? Your model is useless without it.

[TruthIsAll]

10. Useless? Hmm...

tia

[requiem99]

12. Yes, useless. Without the turnout factored in

How can you possibly pretend to predict an accurate result? Turnout is EVERYTHING.

[johan helge]

29. I agree, "useless" isn't right, but I have the same question.

Assume a poll with 1000 respondents has 450 for Kerry, 350 for Bush, and 200 undecided.

Assume 75 % of the undecided vote Kerry.

This can mean that Kerry gets 450 + 200*0.75 = 600 votes.

But one can also assume that e.g. only 80 % of the undecided will vote (while all the 800 others vote). This gives Kerry 450 + 200*0.8*0.75 = 570 votes.

What does your model assume?

And: Thanks for the splendid work!

[endnote]

30. Ok, I want to add that too: thanks for trying.

[AmerDem]

3. Bob Shrumm predicted that Kerry will receive

at least 300 electoral college votes. This was on Meet the Press in the final minutes of the show. Melman refused to give his prediction for shrub.

[papau]

4. What is the result with 55% of undecided?

:-)

[TruthIsAll]

5. For 55%, its 94.54% with 306 EV

tia

[papau]

14. thanks!

:-)

[jean]

6. UV allocation impact on Kerry EV and Win Probability - here

[TruthIsAll]

8. Thanks. I forgot to mention that graph.

tia

[trumad]

11. TruthIsAll

Your relentless work throughout this election is most appreciative...

Thanks a ton,

Tru

[Bush Lies]

22. Thanks

[BlueNomad]

23. good question...

I was thinking the same thing...more like 55-60% IMHO...AND don't forget out undecided clowns have to show up against their undecided clowns...

[TruthIsAll]

25. Correction: for 55%, the probability is 97.1%

tia

[endnote]

13. This is BS. It only encourages Dems to stay home.

[TruthIsAll]

15. On the contrary, they will be motivated to get in on the landslide.

Come on, are YOU less likely to vote because of my post.

Jeez, the stuff that gets thrown around here.

[endnote]

16. Look, say it rains that day. I remember from your post that this is

pretty much a sure thing. So, what the hell. I'm savoring my victory early and stay home, stay warm, and watch TV.
That could be what many people might think. If I know it's 50/50 I have more of an incentive to go out and vote.

[jean]

17. many people don't have the chance to see tia's work - they

depend on the alphabet networks and cable for their info - which is fearfearfear/bush is ahead.

Since you're posting on DU can we assume you're more informed than the average American? We all know the import of this election. So, good try, but you're wrong to want to stifle positive and factual info.

[endnote]

18. "factual" ? what are you talking about ?

There is nothing factual about this analysis...

[TruthIsAll]

26. Nothing factual? Back that statement up with YOUR facts.

MY facts are:

1) the polling data
2) the MOE
3) the normal distribution

My assumption is only that Kerry wins the undecided, and that is based on HISTORICAL FACT that undecideds break for the challenger.

If you want facts, just look at the data.
But you would rather blow smoke.

It's because of guys like you that I did this model. To cut through the fog of ignorance.

Let's see YOUR model.

[endnote]

27. TIA, I don't need to obsess about EV Monte Carlo simulations to

keep my anxieties down. Furthermore, models in this area only have the appearance of objectivity. For each "fact" you have you also have 5 assumptions. That is why I think it's BS. That is also why I don't have a model about this (and yes, I do Monte Carlo simulations for a living, in physics). What I do is be a poll watcher and make sure people are not denied the right to vote. So, it seems to me you are the one who's blowing smoke (you accused me of doing that first).

[TruthIsAll]

31. Look, I welcome criticism, but to call the model "useless", shows that

you are uniformed, rude, a troll, or all three.

I believe there's a real chance of a Kerry landslide. And I say this without factoring in turnout, because frankly, I don't know what it will be. Do you?

You apparently feel there could very well be a landslide. So you agree with the model.

Am I missing something here??

Or are you?

[Awsi Dooger]

32. Well, I have my own model, but not foolish enough to claim it's infallible

I appreciate the work, TIA, but you place too much emphasis on polls, especially state polls. The raw numbers aren't everything, merely a useful tool, one of many relevant factors. I learned that in sports betting more than a decade ago. Plus the poll numbers aren't necessarily real, certainly not to the point they can be blindly accepted without reference to recent demograhic/voting tendencies in the state/nation. For instance, we have almost no chance in Colorado. I don't care what any poll from 2004 indicates. It's been a colossal waste of time and resources.

There are any number of studies that look at factors in presidential elections. I have been reluctant to mention them on DU because I would be accused of negativity, but several of the most basic criteria point to Bush. For example, incumbents overwhelmingly succeed if the party has been in power only one term. I think it's 9 of 10 since 1900, the one loss being Carter in 1980. There is a benefit of a doubt by the nation and no urgency to dismiss the incumbent or his party. Republicans have been overwhelming in that scenario, including Ike in '56, Nixon in '72 and Reagan in '84. The 2004 comparisons to Bush 41 in '92 are not valid, not only due to GW's superior approval rating compared to his dad but the fact the GOP had been in power 3 terms then, as opposed to 1 now.

Here is a paper that looks at other factors, including a second quarter GDP economic trend that also favors Bush by wide margin. It was 3.3 final adjustment.(note: there is one glaring flaw in this paper; the statement Delware and Mew Mexico always vote for the winner, ignoring Gore took them both in 2000) http://64.233.161.104/search?q=cache:wO-UOHiDgSUJ:morris.wharton.upenn.edu/forecast/Political/PDFs/JonesEssayIndicatorsModels%2520_30Mar04_.pdf+Randall+J+Jones+Jr.&hl=en&ie=UTF-8

I scan numerous academic and common sense sources like that, in addition to relying on my Excel projections. My debate watching parites are invaluable. People aren't thrilled with John Kerry. Period. Likeability is never emphasized enough on DU. I'm downgrading the undecideds to 62-63% because he simply isn't magnetic enough to capture more. Currently I have Kerry's likelihood at 55%, both numerically and subjectively.

[TruthIsAll]

33. One Big Difference: Bush is Universally Hated. Fact.

Another is that Bush has led us into a premptive war on lies. Fact.

Another is that it is an accepted fact that Bush was NEVER elected in the first place. He stole the election, SCOTUS ratiigies the coup. Fact.

Another is that a signicant 50% of REPUBLICAN newspapers which endorsed Bush in 2000 don't this time. Fact.

Another is that Bush is the FIRST president since Hoover to lose jobs. Fact.

This is the first election ever in which the whole world expects that a candidate will pull out all stops to STEAL. Fact.

This is by far the most corrupt administration in history. Fact.

Bush makes Nixon look like a saint.

I could go on and on. You get the idea.

All facts.

He's toast.

[msongs]

19. and what if these things happen?

we are invaded by martians. an asteroid hits the earth. you decide to take responsibility for your own actions and vote no matter what.

Msongs
Riverside Ca

[endnote]

20. If people did that we would not have Chimpy to start with.

But people are not like that. Bad rain can make many people stay home if they can think of a good excuse, like "oh well, Kerry has a 99% chance of winning anyway".

[undercover_brother]

21. Do you REALLY believe people stay home based on polls this year?

This is not just another lazy year of voting. How many people have no opinion on this election? In the past, politics was a boring topic few casual citizens talked about. This year EVERYONE has a strong opinion.

There is going to be a monster turnout. Be prepared for LONG lines on Nov 2.

[endnote]

28. I HOPE SO. If it is a heavy rain day in some states, things

can change quite a bit.

[lagniappe]

24. Not a chance. I'll bet most DUers think this is one of the

most important elections in our history. Democrats will turn out. You can bank on it.