Newsweek 7/30 poll (after Kerry speech): a 57%-43% Kerry landslide?

[TruthIsAll]

Yes, if you assume 70% allocation of other/undecided/Nader voters to Kerry.

And focus on the post-speech polling.


from:
pollingreport.com

Newsweek Poll conducted by Princeton Survey Research Associates International. July 29-30, 2004. N=1,010 registered voters nationwide. MoE ± 4 (for all registered voters).


"Suppose the election were being held TODAY and you had the following three choices for president and vice president: George W. Bush and Dick Cheney, the Republicans; John Kerry and John Edwards, the Democrats; and Ralph Nader and Peter Camejo, the independent or third party candidates. Who would you vote for?" If none of these: "As of TODAY, do you LEAN more toward Bush and Cheney, the Republicans; Kerry and Edwards, the Democrats; or Nader and Camejo, the third party candidates?"



Bush/Cheney Kerry/Edwards Nader/Camejo Other (vol.)/Undecided
% % % %
ALL 42 49 3 6
Republicans 90 7 1 2
Democrats 8 86 1 5
Independents 39 45 7 9
Men 48 43 3 6
Women 37 53 3 7
18-29 32 51 9 8
30-49 47 46 2 5
50 & older 44 49 1 6
Southern white 54 38 3 5
Non-South, white 47 46 2 5
Non-white 20 65 5 10
Urban 30 63 2 5
Suburban 47 45 2 6
Rural 49 37 4 10

7/30 only:
40 50 3 7 <<<<<< 57% kerry, 43% bush with 70% allocation of Nader/other/undecided to kerry

7/29 only:
45 47 2 6


Trend:
7/8-9/04 44 47 3 6

This forecast model does NOT include the Newsweek numbers:
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=104x2113091

[bluestateguy]

1. Nobody will win the election by that much

No way.

I don't like political science "models". The only models I am interested in are in the Victoria's Secret catalogue.

[yardwork]

3. The Republicans will get a big bounce after their Convention, too.

Nothing matters until Election Day. And we need to make sure that the votes get counted.

[Eric J in MN]

2. We need to keep it going.

1) Volunteer at
www.johnkerry.com

2) Donate at
www.democrats.org

or

www.democracyforamerica.com

or

www.moveon.org

3) Send letters to the editor.

4) Talk to undecided friends.

[burythehatchet]

4. I'm not a statistician (although I like to read actuarial reports)

but I've been calling landslide since before Kerry was the frontrunner in the primaries. That, of course, has zip to do with who gets in office. We live in a political third world banana republic.

[CaTeacher]

5. YES!!!

It is going to be a landslide! Bush needs to start packing his bags NOW!!

[papau]

9. "like to read actuarial reports" - get to a Doctor immediately! :-)

Why is the SOA, Conference, and the AAA now sending the reports out in PDF format?

It is a pain to convert them to RTF format so that you can use them in memos.

Just an old actuary's complaint of the day!

:-)

:party:

:toast:

:-)

[burythehatchet]

15. A former passing interest and required understanding of

actuarial work has turned into a full blown interest because the actuary that works on my risk finance program is a youngish lady, not too pretty, but very fit, and the way she explains Monte Carlo simulations totally sends me into a state of lustful actuarial bliss. (I hope she's not a DUer).

Pretty bizarre, huh.

[TruthIsAll]

16. Here is my Monte Carlo simulation of the election

I ran 1000 simulation trials, based on probabilities of Kerry winning each state, for three cases: Kerry gets 60%, 70%, 80% of the other/undecided voters.

Bottom line: Kerry has a 96% probability of winning the election, assuming the 70% case.

Take a look at the first 10 trial runs, state by state, to sum up the electoral votes for Kerry.


http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=104x2113091

[papau]

17. What software did you use to do the simulation?

I think I need some new toys!

[TruthIsAll]

18. I used basic Excel: Cum. Normal distribution and Randon number generator.

One major advantage to this approach: The model is not affected by slight movements in state polling numbers, which is the case with Net Blogs who have web sites for forecasting the election.

For example, if the latest poll has Kerry ahead 51%-49% over Bush in FL, he has a 60% probability of winning the state.

Other predictors would give him the full 27 electoral votes.

If the next day, Bush has a 51% lead, he gets all the votes. This causes a dramatic change in the total EV. That is not realistic.

The simulation removes this bias, by running 1000 simulation trials. In the first case, Kerry would win about 600 times; in the second, Bush would win about 600 trials.

The average EV total would not change much between the two runs.

[burythehatchet]

28. Thanks. I'll repeat some conventional wisdom

In looking at those numbers, it's apparent that this election will likely be decided in Ohio...Columbus, specifically. Can someone remind me in which state the voting machine guy promised to deliver the state to idiot boy? :)

Oh, and Truth, that was a nice analysis and all, but you're not doin' it for me the way MY actuary does it for me. (thank God).

[Toots]

6. In 1980 after listening to Reagan I said to myself that he was

going to win by a landslide. He did. I have not said that to myself since then but after listening to Kerry and his daughters I sense the exact same thing. Kerry hit all the right notes.

[yardwork]

7. The problem is still older white rural and suburban men

Those are the categories in which * still leads.

Work on your middle-aged and elderly white male neighbors, especially in the south and midwest.

Bush/Cheney Kerry/Edwards Nader/Camejo Other (vol.)/Undecided
% % % %
ALL 42 49 3 6

Men 48 43 3 6

30-49 47 46 2 5
50 & older 44 49 1 6

Southern white 54 38 3 5
Non-South, white 47 46 2 5

Suburban 47 45 2 6
Rural 49 37 4 10

[blm]

14. Wait till they find out Bush went AWOL. It WILL all come out

and it will be used. Too many still have no idea because of the way the media danced around it. But, there are more cards to be turned over on this one. If the media doesn't do its job the DNC and groups like MoveOn will buy the ads that will do it for them.

[Cat Atomic]

8. How much did Reagan win by in his "landslide", anyway?

It'll be interesting to see how the media describes it in November. I somehow doubt that Kerry's win would be described as a "landslide", even if he won by a larger margin than Reagan did.

You know- just how they kept calling Bush a "popular president" when his approval numbers were 50%.

[burythehatchet]

10. Reagan won 49 states

[yardwork]

12. True, but Reagan didn't win by a huge popular vote margin

It was an electoral college landslide, not a popular vote landslide.

[TruthIsAll]

13. Not vs. Carter in 1980; he won 49 states vs.. Mondale in 1984

tia

Reagan won with 49% to Carter's 41%.

Anderson (Ind.) got 7%

[TruthIsAll]

11. My 7/30 forecast model does NOT include the Newsweek numbers

but DOES include the latest Zogby (before the Kerry speech):


http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=104x2113091

[Zynx]

19. In the end, Kerry will win 53-45 or maybe 52-46 and crush Bush.

[TruthIsAll]

20. I agree with your numbers. I say it will be 54-45-1.

Kerry will win 360 EV.

[Zynx]

29. That could be too, though maybe eight points instead of nine.

I think he will win somewhere between 350ish and 380ish depending on how certain states swing.

[dumpster_baby]

21. white males 30-49 are GOP base, esp. in the South

White males 30-49 are basically the highest earning demographic group, so they are voting in what they see as their own best interest. However, they justify it with all sorts of other reasons.

[ronnykmarshall]

22. WOW!

This is good news.

BUT! The fight must keep on!

[NYCGirl]

23. Don't get too drunk on the numbers just now...

Remember, Dukakis lead by 22 points after the 1988 Democratic convention.

There's still lots of work to do!

[fujiyama]

24. Polls are great when your up

but I always remember that fact.

The Bushies have all sorts of tricks up their sleeves. We don't know the kind of shit they'll pull between now and election day.

But Kerry is in good shape. He seems to have learned from the mistakes of Dukakis and Gore atleast in some ways.

[abrownz]

25. WOOT WOOT n/t

[fujiyama]

26. Did you ever assume a situation

where only 50% of the undecideds would go to Kerry?

[TruthIsAll]

27. Yes. Kerry still wins. If he's ahead 50-40, a 50/50 split makes it 55-45

Lets be realistic. A 50/50 split is way too conservative.

Historically, 60-80% of challengers get the undecided vote.

In this case, Nader voters will abandon him for Kerry.
Undecided independents will go for Kerry.

There's a reason undecided voters are not for Bush: they don't like him. If they did, they would not be undecided.

They want Kerry to close the sale.

He did just that on Thursday night.

[Zynx]

30. 70% of undecideds polled in an NBC/WSJ poll disapproved of Bush

Why would they vote for him?

[buycitgo]

31. mcgovern got the lowest undecideds since truman, right?

sixty percent

what does it come out with kerry getting that?

[TruthIsAll]

Here is sensitivity analysis from the model (see link)

ELECTORAL VOTE WIN PROBABILITY SENSITIVITY ANALYSIS				
One thousand  (1000) trials are run in each simulation.				
				
Simulation I: Conservative Case 				
Assume:	60%	of undecided/other to Kerry.		
				
Kerry:	51.63%	of the vote.		
Wins:	88.8%	of the trials. 		
Avg:	306	electoral votes.		
Max:	386	electoral votes.		
				
			
Simulation II: Most Likely Case 			
Assume:	70%	of undecided/other to Kerry.	
			
Kerry:	52.56%	of the vote.	
Wins:	96.7%	of the trials. 	
Avg:	324	electoral votes.	
Max:	404	electoral votes.	
			
			
Simulation III: Optimistic Case 			
Assume:	80%	of undecided/other to Kerry.	
			
Kerry:	53.48%	of the vote.	
Wins:	99.6%	of the trials. 	
Avg:	340	electoral votes.	
Max:	434	electoral votes.		
				

[TruthIsAll]

32. Here is sensitivity analysis from the model (see link)

ELECTORAL VOTE WIN PROBABILITY SENSITIVITY ANALYSIS				
One thousand  (1000) trials are run in each simulation.				
				
Simulation I: Conservative Case 				
Assume:	60%	of undecided/other to Kerry.		
				
Kerry:	51.63%	of the vote.		
Wins:	88.8%	of the trials. 		
Avg:	306	electoral votes.		
Max:	386	electoral votes.		
				
			
Simulation II: Most Likely Case 			
Assume:	70%	of undecided/other to Kerry.	
			
Kerry:	52.56%	of the vote.	
Wins:	96.7%	of the trials. 	
Avg:	324	electoral votes.	
Max:	404	electoral votes.	
			
			
Simulation III: Optimistic Case 			
Assume:	80%	of undecided/other to Kerry.	
			
Kerry:	53.48%	of the vote.	
Wins:	99.6%	of the trials. 	
Avg:	340	electoral votes.	
Max:	434	electoral votes.		
				

[TruthIsAll]

33. The media whores are minimizing the Kerry bounce.

They fail to mention that half the poll was taken BEFORE the speech.

The post-speech Kerry lead of 50-40 with the majority of the other 10 leaning to Kerry is a harbinger of things to come.

The media is running scared.