8/4 KERRY WIN PROBABILITY: 96.65% ; VOTE: 52.78% ; EV:322

[TruthIsAll]

								
ELECTION FORECAST / SIMULATION MODEL								
Created by TRUTHISALL								
								
National Polling data: Pollingreport.com								
State Polling data: Electoral-vote.com 								
								
COMBINED NATIONAL/STATE MODEL 
Assumes 70% of undecided/other to Kerry

………	Nat.	State	Comb.	Comb.	Win Probability		
………	Model	Model	Vote%	EV	Nat.	State	Comb.
Kerry:	52.88	 52.68 	52.78	322	 96.60 	96.70	96.65
Bush:	47.12	 47.32 	47.22	216	 3.40 	 3.30 	3.35
Spread:	5.76	5.35	5.56	106		
	


LATEST NATIONAL POLLS, FORECAST, WIN PROBABILITY					
15Poll	Avg	Proj.	Prob.
Kerry:	47.93	52.88	 96.60 		
Bush:	45.00	47.12	 3.40 		
Spread:	2.93	5.76			
					
Kerry:	96.6%	probability minimum 50% popular vote.			


NATIONAL VOTE/ WIN PROBABILITY SENSITIVITY ANALYSIS					
Effect of various undecided/other allocation assumptions					
	Assume Kerry allocation of:				
Alloc:	50%	60%	70%	80%	90%
Vote:	51.47	52.17	52.88	53.59	54.29
Prob:	 82.33 	 91.56 	 96.60 	 98.85 	 99.68 

						
STATE FORECAST SIMULATION / WIN PROBABILITY						
Forecast  %    EV          						
Kerry:	52.68	322				
Bush:	47.32	216				
Spread:	5.35	106				
						
Kerry:	96.7%	probability at least 270 electoral votes.				
based on winning		967	of 1000 election trial simulations. 			
						
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:	51.79%	of the vote.	
Wins:	90.3%	of the trials. 	
Avg:	307	electoral votes.	
Max:	389	electoral votes.	
			
			
Simulation II: Most Likely Case 			
Assume:	70%	of undecided/other to Kerry.	
			
Kerry:	52.68%	of the vote.	
Wins:	96.7%	of the trials. 	
Avg:	322	electoral votes.	
Max:	425	electoral votes.	
			
			
Simulation III: Optimistic Case 			
Assume:	80%	of undecided/other to Kerry.		
				
Kerry:	53.57%	of the vote.		
Wins:	99.3%	of the trials. 		
Avg:	337	electoral votes.		
Max:	430	electoral votes.		
				
				


LAST THREE PRESIDENTIAL ELECTIONS				
(in millions of votes)				
Dem	138.78	52.61%		
Rep	125.03	47.39%		

				

BUSH JOB APPROVAL												
The data source is PollingReport.com    												
												
2004	Avg	Nwk	Fox	CNN	Pew	Harris	CBS	ABC	Time	NBC	AP	Zogby
Jan.	54.50	50	58	60	56	na	50	58	54	54	56	49
Feb.	49.67	48	48	51	48	51	50	50	54	na	47	na
Mar.	48.75	48	48	49	46	na	51	50	na	50	48	na
Apr.	48.80	49	50	52	48	48	46	51	49	na	48	47
May	45.20	42	48	47	44	na	41	47	46	47	48	42
June	47.10	48	48	49	48	50	42	47	na	45	48	46
July	47.20	45	47	47	46	na	45	50	50	48	50	44

												
												
												
						
NATIONAL POPULAR VOTE TREND						
Based on latest national polls from:						
IBD, ABC, NWK, ARG, CBS, PEW, LAT, ZOGBY, TIME						
Data source: PollingReport.com    						
						
						
Kerry projection is the current average of these polls						
plus 70% of undecided/other voters.						
						
	Avg Poll Trend		Projection			
	Kerry	Bush 	Diff	Kerry	Bush 	Diff
Jan 	44.0	49.4	-12.7	46.4	53.6	-7.1
Feb	48.0	45.4	2.6	52.6	47.4	5.2
Mar	48.6	44.0	4.6	53.8	46.2	7.5
Apr	47.4	44.9	2.5	52.8	47.2	5.6
May	47.8	44.0	3.8	53.5	46.5	7.1							
June	47.1	44.7	2.4	52.8	47.2	5.7							
July	48.1	45.0	3.1	52.9	47.1	5.9							
Aug	50.0	44.2	5.8	54.1	45.9	8.1							
													
KERRY NATIONAL POPULAR VOTE FORECAST/WIN PROBABILITIES
													
Uses latest national poll (15) data													
Avg 	47.93%	 Latest 15 poll average (mean)											
+Alloc.	4.95%	+ 70% undecided/other 											
= Proj.	52.88%	 = Projected Kerry %											
													
Prob: 	96.60%	> 50% of popular vote											
													
-------	Date			Kerry%	Kerry% undecided/other 		
-------	(mdd)	Kerry	Bush	vs.Bush	50%	60%	70%	80%	90%	
												
TIME	722	50.0	45.0	52.6	52.5	53.0	53.5	54.0	54.5
FOX	721	44.0	43.0	50.6	50.5	51.8	53.1	54.4	55.7
CNN/GAL	801	47.0	50.0	48.5	48.5	48.8	49.1	49.4	49.7
LAT	721	48.0	46.0	51.1	51.0	51.6	52.2	52.8	53.4
PEW	718	46.0	44.0	51.1	51.0	52.0	53.0	54.0	55.0
									
IBD	724	46.0	43.0	51.7	51.5	52.6	53.7	54.8	55.9
CBS	730	49.0	43.0	53.3	53.0	53.8	54.6	55.4	56.2
DEMC	713	52.0	45.0	53.6	53.5	53.8	54.1	54.4	54.7
ABC/WP	802	52.0	45.0	53.6	53.5	53.8	54.1	54.4	54.7
NWK	730	52.0	44.0	54.2	54.0	54.4	54.8	55.2	55.6
									
ZOGBY	729	48.0	43.0	52.7	52.5	53.4	54.3	55.2	56.1
AP	707	45.0	49.0	47.9	48.0	48.6	49.2	49.8	50.4
NBC/WSJ	721	45.0	47.0	48.9	49.0	49.8	50.6	51.4	52.2
ARG	801	49.0	45.0	52.1	52.0	52.6	53.2	53.8	54.4
QPAC	722	46.0	43.0	51.7	51.5	52.6	53.7	54.8	55.9	
										
Mean		47.93	45.00	51.58	51.47	52.17	52.88	53.59	54.29	
Std		2.69	2.20	54.92	1.85	1.81	1.83	1.91	2.04	
StdMean		1.73	1.68	1.58	1.58	1.58	1.58	1.58	1.58	
										
Prob>x		58.10	97.50	15.90	82.33	91.56	96.60	98.85	99.68	
										
95% Confidence Interval										
Max		51.33	48.29	54.68	54.56	55.27	55.97	56.68	57.38	
Min		44.54	41.71	48.48	48.37	49.08	49.79	50.50	51.21	
x		48.29	48.29	50.00	50.00	50.00	50.00	50.00	50.00	
Prob>x		58.10	97.50	15.90	82.33	91.56	96.60	98.85	99.68
 	

99% Confidence Interval										
Max		52.40	49.33	55.66						
Min		43.47	40.67	47.50					
x		49.33	49.33	50.00					
Prob>x		78.96	99.51	15.90					
									
									
Probability:									
99.0%	Each within Max-Min Range								
99.5%	Kerry exceeds Bush Max								
15.9%	Bush exceeds his Max								
									
Notes:									
Std = Standard Deviation (variability)									
Stdm = Standard Error of the Mean 									
									
95% Confidence Limits formula:									
Upper Limit = Mean + 1.96 * Stdm									
Lower Limit = Mean - 1.96 * Stdm								
								
99% Confidence Limits formula:								
Upper Limit = Mean + 2.58 * Stdm								
Lower Limit = Mean - 2.58 * Stdm								
								
								
								
								
								
								
								
ELECTORAL VOTE SIMULATION FORECAST METHODOLOGY								

Kerry National and State Percent Projections								
State vote: Latest state poll plus assumed share of
undecided/other voters.								
National vote: Accumulated state projected votes weighted by
ratio of state/national votes for last three
elections.													
Probability of state win: based on projected vote
%.													


Most Likely Case													
Assume 70% of undecided/other voters for Kerry													
Win	Prob.	Pct	EV										
Kerry	96.7%	52.7%	322										
Bush	3.3%	47.3%	216										

Dem Historical Vote %, Kerry Projection and Win Probability
													
Dem Hist. is Democratic share of 2-party vote in last three
presidential elections.													
													
-------	Dem	Kerry	Kerry	EV Trial runs 1-10 of 1000									
-------	Hist.	Proj.	Prob.	1	2	3	4	5	6	7	8	9	10
State	52.60%	52.68%	96.7%	265	325	275	370	296	307	317	342	316	277
													
AL	44.8	41.0	1.2										
AK	37.6	40.7	1.0										
AZ	48.8	49.9	49.0				10	10					10
AR	55.2	50.2	52.0	6				6		6		6	6
CA	57.4	58.1	97.9		55	55	55	55	55	55	55	55	55
													
CO	48.8	49.3	43.1	9						9			
CT	57.7	62.6	99.9	7	7	7	7	7	7	7	7	7	7
DE	56.8	57.1	96.2	3	3	3	3	3	3	3	3	3	3
DC	90.3	90.4	100.0	3	3	3	3	3	3	3	3	3	3
FL	50.7	52.1	70.0	27	27		27		27		27	27	27
													
GA	47.6	45.7	14.1										15
HI	59.0	60.9	99.7	4	4	4	4	4	4	4	4	4	4
ID	35.7	39.0	0.3										
IL	57.9	60.0	99.4	21	21	21	21	21	21	21	21	21	21
IN	44.8	45.9	15.3										
													
IA	51.8	52.8	75.8		7	7		7	7			7	
KS	42.4	41.6	1.8			6					6		
KY	46.7	46.2	17.1	8									
LA	49.2	44.4	8.1				9						
ME	57.1	52.3	71.7	4		4	4	4	4	4	4		4
													
MD	57.8	58.6	98.4	10	10	10	10	10	10	10	10	10	10
MA	65.3	65.8	100.0	12	12	12	12	12	12	12	12	12	12
MI	54.7	54.1	84.7	17		17	17		17	17	17	17	
MN	55.7	54.8	88.5	10	10		10	10	10	10	10	10	10
MS	44.3	36.3	0.0										
													
MO	52.5	51.1	60.8	11	11	11	11		11	11	11		
MT	44.9	42.8	3.6										
NE	37.5	36.5	0.0										
NV	49.9	51.3	62.7	5	5		5	5			5	5	5
NH	51.7	55.9	93.0	4	4	4	4	4	4	4	4	4	4
													
NJ	56.5	57.0	96.0	15	15	15	15	15	15	15	15	15	15
NM	53.0	51.1	60.8				5		5	5	5	5	
NY	62.6	65.0	100.0	31	31	31	31	31	31	31	31	31	31
NC	46.6	47.5	26.6	15	15		15	15					
ND	40.8	37.2	0.1										
													
OH	50.8	48.1	31.7				20			20	20		
OK	42.8	39.2	0.3										
OR	53.6	52.8	75.8	7	7	7	7	7	7	7	7	7	7
PA	54.2	57.2	96.4		21	21	21	21	21	21	21	21	
RI	65.6	67.2	100.0	4	4	4	4	4	4	4	4	4	4
													
SC	44.4	47.5	26.6	8		8					8		
SD	44.5	44.8	9.7										
TN	50.5	50.2	52.0		11	11	11		11	11			11
TX	44.3	42.6	3.2										
UT	33.6	29.7	0.0										
													
VT	59.4	60.1	99.4	3	3	3	3	3	3	3	3	3	3
VA	47.3	49.9	49.0		13			13		13	13	13	
WA	55.9	53.8	82.9	11	11	11	11	11		11	11	11	
WV	54.0	53.6	81.6		5		5	5	5		5	5	
WI	52.7	51.4	63.7	10	10		10	10	10			10	10
WY	38.3	30.8	0.0										
													
Avg	52.60%	52.68%	96.7%	265	325	275	370	296	307	317	342	316	277

[sundancekid]

1. NICE work, and THANK YOU for posting!

:yourock:

[Hepburn]

2. Wow! n/t

[TruthIsAll]

3. No link; its my own Excel model. Check data links provided.

tia

[Bucky]

4. No. It's probability that polls are right IF the election were held today.

Obviously the election isn't being held today. Your analysis, while getting the math right, still answers a question other than the question you claim that it's answering.

There are three months of X variables and unknown inputs between now and the election. You certainly haven't forecasted the probability of a Kerry victory in November here. You've forecasted the probability that the whole country right at this particular moment, before Bush has had his convention feels the way an aggregation of these polls would tend to show.

That's what we in the market research industry call "bad math" or "misleading the client."

[EstimatedProphet]

5. Well, sure it is

He never said otherwise.
It's still good news. What were the numbers a year ago?

[TruthIsAll]

6. No, its called countering the mediawhore spin that its Tied

Kerry is ahead. Solidly so. Right now. End of Story.

[cthrumatrix]

16. BINGO

[TruthIsAll]

7. I am misleading no one, sir. If you can model in those

unknowns, please do so.

I run this model every day as I get new data. It is not cast in stone.

Don't you do the same with your marketing surveys?

Since you are in the market research industry, show us how to factor in those unknown variables you refer to.

Apparently, you seem to feel that neither Bush or Kerry uses polling for forecasting and strategic planning.

BTW, Bush will get a minimal bounce, if any, at his convention. In fact, I predict his numbers on Sept 1 will be even lower than they are today.

And next time, tell me something I don't know about forecast modeling.

As one with ZERO experience in marketing, I am not as familiar as you may be with the art of misleading the client.

Remember that bogus Hill & Knowlton Gulf War fiction about those mean old Saddam Republican Guards throwing babies out of incubators?

That, my friend, was Bush I misleading the world at the U.N.

The same thing is going on today. The mediawhores at CNN amd FOX claim that Kerry received a NEGATIVE post-convention bounce. Pure Bushit.

My numbers show Bush is stuck at 44% and he can't get up.

[Bucky]

13. Then say THAT instead of "Kerry WIN probability". Truth in Advertising

You have measures of current popularity, not a measure of the outcome in November, which is what your title claims. It's misleading to people who don't understand what statistics are supposed to measure. These numbers you're running have no predictive value. Before you can make a true prediction, you have to factor in other variables, yet we don't know what they are yet.

Since you are in the market research industry, show us how to factor in those unknown variables you refer to

Dude, they're called unknown for a reason.

I run this model every day as I get new data. It is not cast in stone.

Thank your for that caveat. You need to put that in your original post.

Don't you do the same with your marketing surveys?

Good question. No, what we offer to clients is a "snap shot" of where their market share is at a given moment. Of course for shoppers and retailers, every day is "election day" so the stats we generate have some concrete market value. We also couple the data with historical data to produce trends for predicting future performance.

Apparently, you seem to feel that neither Bush or Kerry uses polling for forecasting and strategic planning.

I'm not sure why you say that. To the contrary, I imagine that they do, but that they use numbers besides these simple "who would you vote for if you were voting today?" statistics. They look at below the surface questions about trust, depth of support, possible swing issues, and historical trends in order to assess future outcomes.
And then they label the data correctly when they publish the numbers.

BTW, Bush will get a minimal bounce, if any, at his convention. In fact, I predict his numbers on Sept 1 will be even lower than they are today.

I like what that prediction says. I want it to be true. But what data is it based on?

[Old and In the Way]

8. TIA has always said this is a model based on current data.

No one can predict the future, but this does give us insight into the voting trends and subtle changes that will occur between now and the election.

[Old and In the Way]

9. TIA, you should ask the administrators to set up a forum to post

your analysis weekly. It would be interesting to monitor your probability changes over time.

[TruthIsAll]

10. You can always search the archives; they are a permanent record.

Thanks,

tia

[Old and In the Way]

11. True, but I was thinking it would be a good feature for the DU to

promote. It might bring new readers to the site if your predictive analysis was given some visibility.

[mrboba1]

12. Question about undecideds:

How do you settle on the 70% of undecideds going for the challenger? Is there some history behind that?

I assume there is, I'd like to see it...

tia, tia
;)

[TruthIsAll]

14. The Incumbent Rule

On the other hand, undecideds are ready to committ to the comparatively unknown challenger, who they are just getting to know. They just want him to close the sale. If he does, he gets their vote.

John Kerry and John Edwards are closing the sale.

The Incumbent Rule
http://www.pollingreport.com/incumbent.htm

How will undecideds vote on election day? Traditionally, there have been two schools of thought about how undecideds in trial heat match-ups will divide up at the ballot box. One is that they will break equally; the other, that they will split in proportion to poll respondents who stated a candidate preference.

But our analysis of 155 polls reveals that, in races that include an incumbent, the traditional answers are wrong. Over 80% of the time, most or all of the undecideds voted for the challenger.

The 155 polls we collected and analyzed were the final polls conducted in each particular race; most were completed within two weeks of election day. They cover both general and primary elections, and Democratic and Republican incumbents. They are predominantly from statewide races, with a few U.S. House, mayoral and countywide contests thrown in. Most are from the 1986 and 1988 elections, although a few stretch back to the 1970s.

The polls we studied included our own surveys, polls provided to us directly by CBS, Gallup, Gordon S. Black Corp., Market Opinion Research, Tarrance Associates, and Mason-Dixon Opinion Research, as well as polls that appeared in The Polling Report.

In 127 cases out of 155, most or all of the undecideds went for the challenger:


DISPOSITION OF UNDECIDED VOTERS
.

Most to challenger 127
Split equally 9
Most to incumbent 19



more...

[cthrumatrix]

15. but..but.... bunnypants says he can only keep us safe from a boogeyman