8/15 KERRY: 99.0% PROBABILITY, 52.70% VOTE, 337EV

[TruthIsAll]

ELECTION FORECAST / SIMULATION MODEL								
Created by TRUTHISALL								
								
National Polling data: Pollingreport.com								
State Polling data: electoral-vote.com 								
								
								
PROLOGUE: LAST THREE PRESIDENTIAL ELECTIONS								
(in millions of votes)								
Dem	138.78	52.61%						
Rep	125.03	47.39%						
								
								
NATIONAL / STATE POLLS: AVERAGES AND PROJECTIONS (%)								
National: based on 11-poll group average.							
State: Historical voting weights applied to state polls for
national pct.							
							
Current: undecided/others allocated in proportion to
polls.							
Projection: 60% of undecided/other voter allocated to
Kerry.							
							
National Win Probability: popular vote majority.							
EV Win Prob: Pct. of Kerry EV wins in 1000 simulation trials. 
     							
							
Avg EV Sim: Average Kerry EV in 1000 simulated election
trials.       							
							
______	Curr. 	Curr. 	Curr. 	Proj. 	Proj. 	Proj.	AvgSim
______	Nat.	State	Comb.	Nat.	State	Comb.	ElecVote
Kerry	49.89	46.24	48.07	53.56	51.85	52.70	337
Bush	44.00	44.42	44.21	46.44	48.15	47.30	201
Diff	5.89	1.83	3.86	7.11	3.70	5.40	136
Other	6.11	9.34	7.73				
							
Kerry Win							
Prob	96.69	99.10	97.90	98.14	99.90	99.02	99.90
							
							
NATIONAL POLLS: PROBABILITY SENSITIVITY ANALYSIS							
Based on 15-poll group average							
							
Kerry	Undecided/ other allocations						
Alloc.	50%	55%	60%	65%	70%		
Vote%	52.10	52.44	52.77	53.11	53.45		
Prob%	89.11	92.36	94.82	96.60	97.84		
							
							
STATE POLLS: ELECTORAL VOTE SIMULATION							
One thousand election trials runs in each simulation.			
			
Simulation I: Conservative Case 			
Assume:	50%	of undecided/other to Kerry.	
Kerry	50.91%	of the vote.	
Wins 	99.1%	of 1000 election trials. 	
Avg  	320	electoral votes.	
Max  	387	electoral votes.	
			
Simulation II: Most Likely Case 			
Assume:	60%	of undecided/other to Kerry.	
Kerry	51.85%	of the vote.	
Wins 	99.9%	of 1000 election trials. 	
Avg  	337	electoral votes.	
Max  	413	electoral votes.	
			
Simulation III: Optimistic Case 													
Assume:	70%	of undecided/other to Kerry.											
Kerry	52.78%	of the vote.											
Wins 	100.0%	of 1000 election trials. 											
Avg  	353	electoral votes.											
Max  	398	electoral votes.											
													

BUSH JOB APPROVAL													
Data source: PollingReport.com    													
													
2004	Avg	Nwk	Fox	CNN	Pew	Harris	CBS	ABC	Time	NBC	AP	Zogby	Qpac
Jan	54.50	50	58	60	56	na	50	58	54	54	56	49	53
Feb	49.67	48	48	51	48	51	50	50	54	na	47	na	48
Mar	48.75	48	48	49	46	na	51	50	na	50	48	na	49
Apr	48.55	49	50	52	48	48	46	51	49	na	48	47	46
May	45.18	42	48	47	44	na	41	47	46	47	48	42	45
June	47.00	na	48	49	48	50	42	47	na	45	48	46	na
July	47.82	48	47	47	46	na	45	50	50	48	50	49	46
Aug	46.25	45	44	51	46	na	44	47	na	na	49	44	na
													
													
													
													
NATIONAL POLLS:  MONTHLY TREND / PROJECTION													
Data source: PollingReport.com    													
													
Based on average of 11-poll group:													
IBD, ABC, NWK, ARG, CBS, PEW, LAT, ZOGBY, TIME,DEMCORP,
QPAC													
													
NBC, CNN, AP and FOX are EXCLUDED from this
group.													
These are INCLUDED in the 15-Poll group.													
									
Kerry is at an ALL-TIME HIGH of 49.67%									
Bush is at an ALL-TIME LOW of 44.00%									
									
Curr. Kerry is equal to Kerry/(Kerry+Bush).									
Proj. Kerry is equal to Kerry plus 60% of
undecided/other.									
									
National Trend---------------	Curr	Curr	Proj.	Proj.	
2004	Kerry	Bush 	Other	Diff	Kerry	Prob	Kerry	Prob.	
Jan 	44.8	48.7	6.5	-3.8	48.0	11.5	48.7	22.9	
Feb	48.7	45.3	6.0	3.3	51.8	85.1	52.3	90.8	
Mar	48.1	44.6	7.3	3.6	51.9	86.9	52.5	92.9	
Apr	47.1	45.1	7.8	2.0	51.1	73.7	51.8	85.1	
May	47.6	44.4	8.0	3.2	51.7	84.6	52.4	92.0	
June	47.4	45.1	7.5	2.3	51.2	76.2	51.9	86.4	
July	48.2	44.9	6.9	3.3	51.8	84.8	52.3	91.4	
Aug	49.89	44.00	6.11	5.89	53.14	96.69	53.56	98.14					

													
													
													
NATIONAL FORECAST SCENARIOS 													
Based on 15-poll Group													
													
48.73%	Average (mean) of 15 polls 												
4.04%	plus 60% of undecided/other voters												
52.77%	Projected Kerry vote												
-------------													
94.82%	Probability of a popular vote majority												
													
Nat.	Poll	Poll	Curr. 	Curr. 	Curr	Undecided/other % alloc.
							
Poll	Date	MoE	Kerry	Bush	K/K+B	50.0	55.0	60.0	65.0	70.0			
										
TIME	805	3.00	51	43	54.3	54.0	54.3	54.6	54.9	55.2
FOX	804	4.00	46	43	51.7	51.5	52.1	52.6	53.2	53.7
CNN	811	4.00	47	48	49.5	49.5	49.8	50.0	50.3	50.5
LAT	721	3.00	48	46	51.1	51.0	51.3	51.6	51.9	52.2
PEW	810	3.50	47	45	51.1	51.0	51.4	51.8	52.2	52.6
										
IBD	805	3.50	49	43	53.3	53.0	53.4	53.8	54.2	54.6
CBS	730	3.00	49	43	53.3	53.0	53.4	53.8	54.2	54.6
DEMC	805	3.10	52	45	53.6	53.5	53.7	53.8	54.0	54.1
ABC	802	3.00	52	45	53.6	53.5	53.7	53.8	54.0	54.1
NWK	730	4.00	52	44	54.2	54.0	54.2	54.4	54.6	54.8
										
ZOGBY	814	3.10	50	43	53.8	53.5	53.9	54.2	54.6	54.9
AP	806	3.50	48	45	51.6	51.5	51.9	52.2	52.6	52.9
NBC	721	3.40	45	47	48.9	49.0	49.4	49.8	50.2	50.6
ARG	801	3.50	49	45	52.1	52.0	52.3	52.6	52.9	53.2
QPAC	722	2.50	46	43	51.7	51.5	52.1	52.6	53.2	53.7
										
Mean	______	3.34	48.73	44.53	52.25	52.10	52.44	52.77	53.11	53.45
ProbWin	______	______	______	______	90.68	89.11	92.36	94.82	96.60	97.84
										
										
95% Confidence Interval										
Max		______	52.1	47.9	55.6	55.4	55.8	56.1	56.5	56.8
Min		______	45.4	41.2	48.9	48.8	49.1	49.4	49.8	50.1
x		______	47.9	47.9	50.0	50.0	50.0	50.0	50.0	50.0
Prob>x		______	69.3	2.5	90.68	89.11	92.36	94.82	96.60	97.84
										
99% Confidence Interval										
Max		______	53.1	48.9	56.6					
Min		______	44.3	40.1	47.9					
x		______	48.9	48.9	50.0					
Prob>x		______	45.4	0.5	90.7					
										
MoE	3.34									
Std	1.70									
Std = MoE / 1.96 (95% confidence)										
										
95% Confidence Limits:										
Max = Mean + 1.96 * Std										
Min = Mean - 1.96 * Std										
										
99% Confidence Limits:										
Max = Mean + 2.58 * Std										
Min = Mean - 2.58 * Std										
										
										
Avg	11-poll	K/K+B	15-poll	K/K+B					
Kerry	49.89%	53.14%	48.73%	52.25%					
Bush	44.00%	46.86%	44.53%	47.75%					
Diff	5.89%	6.27%	4.20%	4.50%					
									
									
EFFECT OF POLLING SAMPLE SIZE,  MOE AND STD									
ON THE ACCURACY OF THE SAMPLE MEAN									
									
National polls sample about 1000 of 150 mm potential
voters.									
State Polls sample approx. 600 voters									
									
Margin of Error (MoE):									
Is 4.0% for a 600 sample size.									
Is 3.1% for a 1000 sample size.									
Decreases as the polling sample size (n) increases.									
Is 0.80% for 15 polls (equivalent to 15000 samples).									
									
MoE = 1.96 / ( 2*sqrt (n) ) = 1.96* Std									
where the Standard Deviation Std = 1 / ( 2*sqrt( n ))									
									
The Win Probability Matrix displays probabilties for									
various polling sample sizes.									
									
The 11-poll and 15-poll group averags are used 									
to determine the probability of a Kerry majority
vote,									
assuming up to 15,000 samples (15 polls).									
									
The win probabilities are much higher (a fraction below
100%)									
than probabilitiies computed from a typical 1000
sample.									
It makes sense intuitively that if more samples are
taken									
from a population, the more accurate the sample mean.									
									
									
KERRY NATIONAL VOTE WIN PROBABILITY MATRIX									
(based on sample size and percent mean)									
									
Allocation of undecided/other votes to Kerry:									
Current -  proportional to latest poll spread									
Projected - assumes 60% allocation									
									
									
_____	_____	_____	_____	_____	_____	Nat15 	_____	Nat11 	
Sample	Sample	Std				Curr.	Proj.	Curr.	Proj.
Size	MoE	Dev	50.5%	51.0%	52.0%	52.25%	52.77%	53.14%	53.56%
									
______	______	______	KERRY WIN PROBABILITY MATRIX						
State Poll Avg:									
600	4.00%	2.04%	59.7	68.8	83.6	86.50	91.29	93.78	95.92
Nat. Poll Avg:									
858	3.34%	1.71%	61.5	72.1	87.9	90.64	94.79	96.69	98.14
									
1000 	3.10%	1.58%	62.4	73.6	89.7	92.3	96.03	97.63	98.77
1250 	2.77%	1.41%	63.8	76.0	92.1	94.4	97.51	98.67	99.40
1500 	2.53%	1.29%	65.1	78.1	93.9	95.9	98.42	99.24	99.71
									
2000 	2.19%	1.12%	67.3	81.4	96.3	97.8	99.34	99.75	99.93
3000 	1.79%	0.91%	70.8	86.3	98.6	99.3	99.88	99.97	100.00
4000 	1.55%	0.79%	73.6	89.7	99.4	99.8	99.98	100.00	100.00
5000 	1.39%	0.71%	76.0	92.1	99.8	99.9	100.00	100.00	
6000 	1.27%	0.65%	78.1	93.9	99.9	100.0	100.00		
									
7000 	1.17%	0.60%	79.9	95.3	100.0	100.0			
8000 	1.10%	0.56%	81.4	96.3					
9000 	1.03%	0.53%	82.9	97.1					
10000 0.98%	0.50%	84.1	97.7					
11000	0.93%	0.48%	85.3	98.2					
									
12000	0.89%	0.46%	86.3	98.6					
13000	0.86%	0.44%	87.3	98.9					
14000	0.83%	0.42%	88.2	99.1					
15000	0.80%	0.41%	89.0	99.3					
									
									
									
									
									
ELECTORAL VOTE SIMULATION FORECAST METHODOLOGY									

Kerry National and State Vote Projections									
State vote%: Latest state poll plus assumed share of
undecided/other voters.													
National vote%:Weighted sum of projected % state
votes.													
Probability of a state win is based on projected %
vote.													


Most Likely Case: 60% of undecided/other to
Kerry.													
													
Win	Prob.	Pct	EV										
Kerry	99.90%	51.85%	337										
Bush	0.10%	48.15%	201										

Historical Vote %, Kerry Projection and Win Probability
													
Hist. Dem share vs Repub in last three presidential
elections.													
													
______	Dem	Kerry	Kerry	EV (election trials 1-10 of 1000)
									
______	Hist.	Proj.	Prob.	1	2	3	4	5	6	7	8	9	10
State	52.6%	51.85%	99.9%	333	338	344	330	363	307	334	343	382	343
													
AL	44.8	43.0	0.0										
AK	37.6	39.6	0.0										
AZ	48.8	49.2	34.8									10	
AR	55.2	49.6	42.2		6	6		6		6		6	
CA	57.4	56.0	99.8	55	55	55	55	55	55	55	55	55	55
													
CO	48.8	48.4	21.7					9					
CT	57.7	60.8	100.0	7	7	7	7	7	7	7	7	7	7
DE	56.8	56.8	100.0	3	3	3	3	3	3	3	3	3	3
DC	90.3	89.0	100.0	3	3	3	3	3	3	3	3	3	3
FL	50.7	54.2	98.0	27	27	27	27	27	27	27	27	27	27
													
GA	47.6	45.2	0.9										
HI	59.0	54.6	98.8	4	4	4	4	4	4	4	4	4	4
ID	35.7	37.0	0.0										
IL	57.9	58.2	100.0	21	21	21	21	21	21	21	21	21	21
IN	44.8	43.2	0.0										
													
IA	51.8	51.6	78.3		7	7	7		7	7	7	7	7
KS	42.4	40.8	0.0										
KY	46.7	45.6	1.6										
LA	49.2	43.2	0.0										
ME	57.1	52.8	91.5	4	4	4	4	4	4		4	4	4
													
MD	57.8	57.8	100.0	10	10	10	10	10	10	10	10	10	10
MA	65.3	64.4	100.0	12	12	12	12	12	12	12	12	12	12
MI	54.7	54.4	98.4	17	17	17	17	17	17	17	17	17	17
MN	55.7	51.8	81.1	10	10	10		10		10	10	10	10
MS	44.3	35.4	0.0										
													
MO	52.5	50.8	65.2	11	11	11			11	11	11	11	11
MT	44.9	41.4	0.0										
NE	37.5	36.0	0.0										
NV	49.9	50.4	57.8		5				5			5	5
NH	51.7	54.4	98.4	4	4	4	4	4	4	4	4	4	4
													
NJ	56.5	55.6	99.7	15	15	15	15	15	15	15	15	15	15
NM	53.0	54.2	98.0	5	5	5	5	5	5	5	5	5	5
NY	62.6	60.2	100.0	31	31	31	31	31	31	31	31	31	31
NC	46.6	49.2	34.8	15		15		15	15	15	15	15	
ND	40.8	36.6	0.0										
													
OH	50.8	52.2	85.9	20	20	20	20	20		20	20	20	20
OK	42.8	40.8	0.0										
OR	53.6	52.4	88.0	7	7	7	7	7	7	7	7	7	7
PA	54.2	54.6	98.8	21	21	21	21	21	21	21	21	21	21
RI	65.6	64.6	100.0	4	4	4	4	4	4	4	4	4	4
													
SC	44.4	47.0	7.1	8									
SD	44.5	43.4	0.1										
TN	50.5	49.6	42.2			11	11	11			11	11	11
TX	44.3	41.8	0.0										
UT	33.6	28.6	0.0										
													
VT	59.4	58.8	100.0	3	3	3	3	3	3	3	3	3	3
VA	47.3	49.2	34.8				13	13				13	
WA	55.9	54.4	98.4	11	11	11	11	11	11	11	11	11	11
WV	54.0	52.8	91.5	5	5		5	5	5	5	5		5
WI	52.7	51.2	72.2		10		10	10				10	10
WY	38.3	30.4	0.0										
													
Avg	52.6%	51.85%	99.9%	333	338	344	330	363	307	334	343	382	343
						

[Placebo]

1. Just scrolling through that has made me go insane...

:crazy:

eh?

[Stephanie]

2. TIA, what do you say to this guy?

What is this all about?

http://www.nytimes.com/2004/08/15/magazine/15QUESTIONS.html

[gristy]

4. I have been hearing about this guy, and what I think he has done

is essentially "curve-fit" some data (haven't seen what data) to an outcome (Presidential winner). I have this sneaking suspicion that he trolled around for data that seemed to be predictive of which party would win the election. Didn't matter if the data was particularly relevant. Once he had his predictive data, he built his model around it and then used it to "predict" how Bush is going to do. If the "model" predicted that Bush is going to win in a landslide, then he's done and he called his friends at the NY Times. If the "model" predicted that Kerry is going to win, then he trolled his data sets some more. His goal is simply to find a set of data that predict past election results AND a Bush win. Doesn't seem it would be too hard. But it isn't science.

[Stephanie]

6. Thanks TIA - But the guy claims to be for Kerry

But right before that he says that whatever answer he gives on who he'll vote for would skew our perception of the poll. So who knows if he's lying or not? The reporter seemed skeptical about the science of it.

[gristy]

9. He claims to be for Kerry?? He's lying.

Fraudlent analysis showing a Bush win. He's either stupid or he's a Bush supporter. LOL!

[DemocratSinceBirth]

10. I Believe He Believes His Model....

I read the unterview in the New York Times....


If he is a Kerry supporter he is the most conflicted Kerry supporter in the world because a Kerry win will discredit his model and by implication him...

[fujiyama]

7. Funny thing is

the professor claims he is a Kerry supporter.

I don't think his methods reflect any bias in any way. I think his methods are flawed. After all, he also predicted poppy Bush would win in '92.

[TreeHuggingLiberal]

11. From what he stated in the interview...

He says that he bases most of his points on the economy. Which still doesn't explain his picking of poppy in 92, but he is betting everything on the economy being peachy in November. He also appears to have donated to Howard Dean http://www.opensecrets.org/indivs/search.asp?NumOfThou=0&txtName=Fair%2C+Ray&txtState=%28all+states%29&txtZip=&txtEmploy=&txtCand=&txt2004=Y&txt2002=Y&txt2000=Y&Order=N

[ThatPoetGuy]

3. TruthIsAll... you rock. EOM.

[Malikshah]

5. So very very true--- these probability bits are great for

Morale!

Wunnerful wunnerful job!

[DrMath]

8. This is almost as good

Here is an UN-biased analysis:

http://www.spacerad.com/electoral/