TIA Forecast Simulation: Kerry Wins 99 of 100 trials with 52.92%, 330EV

[TruthIsAll]

						

TIA NATIONAL PROJECTION MODEL						
National polling data from pollingreport.com includes:						
IBD, ABC, NWK, ARG, NBC, CBS, PEW, LAT, ZOGBY, TIME						
Kerry proj. to win 53.87% of two-party vote.				

TIA STATE FORECAST SIMULATION MODEL						
State polls data from Electoral-Vote.com 						
Kerry 52.92%	330	EV			
Bush	47.08%	208	EV			
						
Kerry wins 		99	of 100 simulation trial runs.			
Kerry has a 	99%	probability of winning the election			
						
				
ELECTORAL VOTE SIMULATION SUMMARY				
				
Data is downloaded to Excel for the forecasting model.				
One hundred (100)  trials are run in each simulation.				
				
Simulation I: Conservative Case 				
Kerry allocation: 	60%	of undecided/other votes.		
Kerry won 	96	trials with 	52.00%	of the vote.
Kerry averaged	311	electoral votes.		
Maximum EV	371			
Minimum  EV	233			
				
Simulation II: Most Likely Case 				
Kerry allocation: 	70%	of undecided/other votes.		
Kerry won 	99	trials with 	52.92%	of the vote.
Kerry averaged	330	electoral votes.		
Maximum EV	379			
Minimum  EV	265			
				
Simulation III: Optimistic Case 				
Kerry allocation: 	80%	of undecided/other votes.		
Kerry won 	100	trials with 	53.83%	of the vote.
Kerry averaged	349	electoral votes.		
Maximum EV	401			
Minimum  EV	297			
				


LAST THREE PRESIDENTIAL ELECTIONS				
(in millions of votes)				
Dem	138.75	52.60%		
Rep	125.03	47.40%										

												

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	47.00	na	48	49	48	50	42	47	na	45	48	46
July	47.40	48	na	na	na	na	45	na	na	45	50	49
												
						
						
						
						
TIA NATIONAL POLL PROJECTION METHODOLOGY						
						
The data source is PollingReport.com    						
Monthly polling data from: 						
IBD, ABC, NWK, ARG, NBC, CBS, PEW, LAT, ZOGBY, TIME						
						
The projection is based on the latest monthly average of the
ten polls.						
The average is adjusted by allocating  70% of undecided/other
voters to Kerry.						
						
						
						
	Poll Averages			Projection		
	Kerry	Bush 	Diff	Kerry	Bush 	Diff		
Jan 	42.50	50.17	-11.50	46.48	53.52	-7.03		
Feb	48.00	45.43	2.57	52.60	47.40	5.20		
Mar	48.13	44.38	3.75	53.38	46.63	6.75		
Apr	47.38	44.88	2.50	52.80	47.20	5.60		
May	47.11	44.22	2.89	53.18	46.82	6.36		
June	47.13	45.00	2.13	52.64	47.36	5.28		
July	49.57	44.29	5.29	53.87	46.13	7.74		
								
								
Kerry National Poll Projection Formula:								
10-Poll Avg 	49.57							
+ allocation	4.30							
= Projection:	53.87	% of total vote.						
								
								
TIA ELECTORAL VOTE SIMULATION FORECAST METHODOLOGY					

Kerry's vote % is weighted by each state's percentage of the
total vote in the last 3 elections..					

Kerry's projected % for each state is based on the latest
state poll, adjusted					
by the allocation of undecided/other voters.					

The probability of Kerry winning a state is determined by the
current poll spread between Kerry and Bush, compared to the
MoE. 					
Obviously, the greater the spread, the greater the probability
Kerry will win the state.					


Simulation II	70%	Undecided/other to Kerry			
Summary	Wins	Pct	AvgEV	MaxEV	MinEV
Kerry	99	52.92%	330	379	265
Bush	1	47.08%	208	273	159


State	Hist.	Proj.	Prob.	Electoral Vote Simulation Trials 1-10
of 100									
	Dem%	Kerry%	Kerry	1	2	3	4	5	6	7	8	9	10
	52.60%	52.92%	Win	330	356	326	342	367	363	346	349	324	348
													
AL	44.8	45.0%	1.0%										
AK	37.6	40.7%	0.1%										
AZ	48.8	47.2%	16.4%						10				
AR	55.2	50.6%	55.8%	6	6	6		6	6		6	6	6
CA	57.4	57.2%	99.0%	55	55	55	55	55	55	55	55	55	55
													
CO	48.8	49.3%	41.4%	9			9				9		
CT	57.7	62.6%	99.0%	7	7	7	7	7	7	7	7	7	7
DE	56.8	57.1%	99.0%	3	3	3	3	3	3	3	3	3	3
DC	90.3	90.4%	99.0%	3	3	3	3	3	3	3	3	3	3
FL	50.7	53.3%	88.4%	27	27	27	27	27	27	27	27	27	27
													
GA	47.6	45.9%	1.0%										
HI	59.0	60.9%	99.0%	4	4	4	4	4	4	4	4	4	4
ID	35.7	39.0%	0.1%										
IL	57.9	60.0%	99.0%	21	21	21	21	21	21	21	21	21	21
IN	44.8	45.9%	1.0%										
													
IA	51.8	53.6%	95.0%	7	7	7	7	7	7	7	7	7	7
KS	42.4	41.6%	0.1%										
KY	46.7	45.3%	1.0%										
LA	49.2	49.0%	37.5%		9						9		
ME	57.1	52.3%	76.9%	4	4	4	4	4		4	4		4
													
MD	57.8	58.6%	99.0%	10	10	10	10	10	10	10	10	10	10
MA	65.3	66.3%	99.0%	12	12	12	12	12	12	12	12	12	12
MI	54.7	54.2%	97.5%	17	17	17	17	17	17	17	17	17	17
MN	55.7	53.3%	88.4%	10	10	10	10	10	10	10		10	10
MS	44.3	36.3%	0.1%										
													
MO	52.5	52.1%	75.0%		11	11	11	11	11	11	11	11	
MT	44.9	42.8%	0.1%										
NE	37.5	36.5%	0.1%										
NV	49.9	51.4%	65.4%	5					5		5		5
NH	51.7	56.7%	99.0%	4	4	4	4	4	4	4	4	4	4
													
NJ	56.5	56.6%	99.0%	15	15	15	15	15	15	15	15	15	15
NM	53.0	55.3%	99.0%	5	5	5	5	5	5	5	5	5	5
NY	62.6	66.4%	99.0%	31	31	31	31	31	31	31	31	31	31
NC	46.6	48.9%	37.5%		15			15			15		15
ND	40.8	37.2%	0.1%										
													
OH	50.8	51.1%	61.5%				20	20	20	20		20	20
OK	42.8	35.2%	0.1%										
OR	53.6	56.2%	99.0%	7	7	7	7	7	7	7	7	7	7
PA	54.2	54.1%	97.5%	21	21	21	21	21	21	21	21	21	21
RI	65.6	64.2%	99.0%	4	4	4	4	4	4	4	4	4	4
													
SC	44.4	47.5%	20.2%	8									
SD	44.5	44.8%	0.1%										
TN	50.5	50.8%	57.7%	11	11		11	11	11	11			11
TX	44.3	42.6%	0.1%										
UT	33.6	29.7%	0.1%										
													
VT	59.4	60.1%	99.0%	3	3	3	3	3	3	3	3	3	3
VA	47.3	48.5%	31.7%		13	13		13	13	13	13		
WA	55.9	55.1%	99.0%	11	11	11	11	11	11	11	11	11	11
WV	54.0	47.2%	16.4%			5							
WI	52.7	55.0%	99.0%	10	10	10	10	10	10	10	10	10	10
WY	38.3	30.8%	0.1%										
													
Avg	52.60%	52.92%	-	330	356	326	342	367	363	346	349	324	348
													
													
													
													

ELECTORAL VOTE SIMULATION SUMMARY													

Sensitivity		Vote%	AvgEV	MaxEV	MinEV								

Simulation I													
(Conservative)													
Kerry % Und/Oth	60%												
Kerry wins	96	52.00%	311	371	233								
Bush wins	4	48.00%	227	305	167
					
Simulation II					
(Most Likely)					
Kerry % Und/Oth	70%				
Kerry wins	99	52.92%	330	379	265
Bush wins	1	47.08%	208	273	159
					
Simulation III					
(Optimistic)					
Kerry % Und/Oth	80%				
Kerry wins	100	53.83%	349	401	297
Bush wins	0	46.17%	189	241	137




	Polling Average Trend			Projection		
Poll	Kerry	Bush	Diff	Kerry	Bush	Diff

IBD						
Feb	44	41	3	54.50	45.50	9.00
Mar	45	43	2	53.40	46.60	6.80
Apr	40	44	-4	51.20	48.80	2.40
May	43	42	1	53.50	46.50	7.00
June	43	44	-1	52.10	47.90	4.20
July	49	44	5	53.90	46.10	7.80
						
						
ABC						
Feb	52	43	9	55.50	44.50	11.00
Mar	53	44	9	55.10	44.90	10.20
Apr	48	49	-1	50.10	49.90	0.20
May	49	47	2	51.80	48.20	3.60
June	53	45	8	54.40	45.60	8.80
						
						
						
AP						
Jan 	37	54	-17	43.30	56.70	-13.40
Mar	45	46	-1	51.30	48.70	2.60
Apr	44	45	-1	51.70	48.30	3.40
May	43	46	-3	50.70	49.30	1.40
July	45	49	-4	49.20	50.80	-1.60
						
						
NWK						
Jan 	41	52	-11	45.90	54.10	-8.20
Feb	50	45	5	53.50	46.50	7.00
Mar	48	45	3	52.90	47.10	5.80
Apr	50	43	7	54.90	45.10	9.80
May	46	45	1	52.30	47.70	4.60
July	51	45	6	53.80	46.20	7.60
						
						
ARG						
Jan 	47	46	1	51.90	48.10	3.80
Feb	48	46	2	52.20	47.80	4.40
Mar	50	43	7	54.90	45.10	9.80
Apr	50	44	6	54.20	45.80	8.40
May	47	44	3	53.30	46.70	6.60
June	48	46	2	52.20	47.80	4.40
July	49	45	4	53.20	46.80	6.40
						
						
NBC						
Jan 	35	54	-19	42.70	57.30	-14.60
Mar	45	47	-2	50.60	49.40	1.20
May	42	46	-4	50.40	49.60	0.80
June	47	47	0	51.20	48.80	2.40
July	54	43	11	56.10	43.90	12.20
						
						
FOX						
Jan 	32	54	-22	41.80	58.20	-16.40
Feb	43	47	-4	50.00	50.00	0.00
Mar	44	44	0	52.40	47.60	4.80
Apr	42	43	-1	52.50	47.50	5.00
May	42	42	0	53.20	46.80	6.40
June	42	48	-6	49.00	51.00	-2.00
						
						
CBS						
Jan 	48	43	5	54.30	45.70	8.60
Feb	47	46	1	51.90	48.10	3.80
Mar	48	43	5	54.30	45.70	8.60
Apr	48	43	5	54.30	45.70	8.60
May	49	41	8	56.00	44.00	12.00
June	45	44	1	52.70	47.30	5.40
July	49	44	5	53.90	46.10	7.80
						
						
CNN/Gallup						
Jan 	43	55	-12	44.40	55.60	-11.20
Feb	48	49	-1	50.10	49.90	0.20
Mar	52	44	8	54.80	45.20	9.60
Apr	46	51	-5	48.10	51.90	-3.80
May	49	47	2	51.80	48.20	3.60
June	48	49	-1	50.10	49.90	0.20
July	50	46	4	52.80	47.20	5.60
						
						
PEW						
Jan 	41	52	-11	45.90	54.10	-8.20
Feb	47	47	0	51.20	48.80	2.40
Mar	48	44	4	53.60	46.40	7.20
Apr	47	46	1	51.90	48.10	3.80
May	50	45	5	53.50	46.50	7.00
June	46	48	-2	50.20	49.80	0.40
						
						
LA Times						
Apr	49	46	3	52.50	47.50	5.00
June	51	44	7	54.50	45.50	9.00
						
						
						
ZOGBY						
Mar	48	46	2	52.20	47.80	4.40
Apr	47	44	3	53.30	46.70	6.60
May	47	42	5	54.70	45.30	9.40
June	44	42	2	53.80	46.20	7.60
July	46	44	2	53.00	47.00	6.00
						
						
TIME						
Jan	43	54	-11	45.10	54.90	-9.80
Feb	48	50	-2	49.40	50.60	-1.20
May	51	46	5	53.10	46.90	6.20
July	49	45	4	53.20	46.80	6.40
…………………………………………………………………………………………………….						


Notes:						
1) State and National polls are adjusted assuming that						
70% of undecided/Nader/others will split for Kerry.           
                						
2) State polls are weighted by its ratio of total votes cast
in the last 3 elections						
3) Third party candidates not considered for this analysis.   
                           						
4) The analysis is only as good as the quality of polling
data						
5) State polls are taken over weeks; National polls are taken
over days.						


Simulation Methodology:						
Monte Carlo simulation uses random numbers applied to						
probability distributions of uncertain events in order to
generate a range of
sampled outcome scenarios. For the analysis, I used the
latest state polling numbers to simulate an election outcome.

                                
In this case, an "event" is defined as Kerry's
forecast share of a
state's popular vote. An "outcome" is the simulated
probability of Kerry winning the election, based on the total
electoral votes for the states he "wins"..          
                     
                                
For example, if we run 100 trials in the simulation and
Kerry's
electoral vote total exceeds 269 in 90 of the 100
trials, we can reasonably conclude that he has a 90%
probability of winning the election. Each state poll is
assumed to have a
margin of error of +/-4% and a uniform distribution
around the polling numbers.                                   
                       
                                
We assign a probability to Kerry winning each state,
determined by the latest polling numbers.

For example, assume these are the probabilities that Kerry
would win the following states:
Kansas: .001                            
New York: .99                          
Florida: .807                           

This means that if the latest Kansas poll is 60% Bush, 40%
( a 20% spread) then Kerry has a 1 in 1,000 chance of
winning the state. Conversely, he has a 99% chance to win NY.
                        
In the case of a close state, like Florida, a Kerry lead of 5%
assigns him an 80.7% probability of winning the state.        
                      
                                
Excel's random generator function returns a random  
number X between 0 and 1. If X is less than
the probability of Kerry winning the state, the state goes to
Kerry, otherwise it goes to Bush.                             
 
                                
For instance, if X = .72 for Florida, then Kerry wins the
state, since X = .72 is LESS than the .807 probability
that Kerry will win the state.

                                
The simulation runs identically for all states. The electoral
votes are
summed for those states which fall in Kerry's column.         
                 
                                
The model assumes that Kerry's polling  numbers in each state
will increase by 70% of the undecided/other vote. This
assumption is based
on historical data, which show that undecideds break for the
challenger over 80% of the time. .
                                                              
 
For example, assume Kerry is leading in a state by 46%-44%.
with 10% undecided. Allocating 70% of the 10% to Kerry adds 7%
to his total. The adjusted projection is now Kerry winning by
53-47%.

[wryter2000]

1. Thanks and a question

I've been meaning to ask you about "undecideds" going for the challenger. It makes sense, but I was wondering if you have some mathematical model that predicts it or data from past elections that show it. (I believe you. I'm just looking for reassurance. :))

And, at what point do they tend to go over to the challenger? Right after a convention? Just before the election? On election day? When should we see the shift?

I could have asked privately, but I figured others might want to hear your answer. TIA :bounce:

[TruthIsAll]

3. GOP firm says undecideds will go for Kerry

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=132&topic_id=581200

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

[wryter2000]

4. Thanks eom

eom

[delhurgo]

2. You do a great job on this.

I just don't know how valuable it is though considering how much polls can change between now and election day.

I expect your analysis would have shown Dukakis ahead by similar margins as well. But we all know what happened then.

I have to be honest, I just can't put much stock in polls at this point. I do enjoy your analysis though, and like seeing how it changes from month to month.

[TruthIsAll]

5. Will the Kerry trend bring him from 52.92% to 54% on 7/31?

Here are the Kerry projections in the Electoral vote simulation model based on state polling data over the last two weeks. Kerry is up +.67% since 7/12. That's a 1.34% spread increase in just in 5 days.

If Kerry adds 1.08% to his current 52.92% and thus reach 54%, an 8.00% spread over Bush, over the next 14 days, he will be well on his way to a landslide.


Date Kerry Proj.
7/2 52.42%
7/4 52.17%
7/7 52.29%
7/8 52.35%
7/10 52.30%
7/12 52.25%
7/13 52.48%
7/14 52.91%
7/15 52.85%
7/17 52.92%

[Mayberry Machiavelli]

7. The big difference between Shrub I/Dukakis and now, is that shrub II is an

incumbent. This was also not true during the shrub II/Gore race.

The significance being that in such a race, all voters including "undecided" have had a chance to see results of shrub II policies for 3.5 years. Hence those that are not already in the shrub II camp will generally not go there (i.e. they have already decided he sucks).

[merh]

6. I'd like to think this is accurate, but I have my doubts....

MS 44.3 36.3% 0.1%

I don't see the dems taking Mississippi.

[TruthIsAll]

8. You are reading the data wrong.

44.1% is the Dems % of the total MS vote (Dems and Repubs only) in the last 3 elections.

36.1% is Kerry's projected MS vote vs. Bush 63.9%, based on the latest polls.

.1% is the Probability of Kerry carrying the state - a miniscule 1 out of a thousand chance.

Look at the trial runs, you will not see a Kerry win in MS.

[merh]

9. Thank you for clarifying that for me.

That makes sense, I know MS is not going to be a Kerry win. :cry:

[TruthIsAll]

10. Kick - no update

.

[Jack Rabbit]

11. Very encouraging

While you can't vouch for the accuracy of the state-by-state data, their summation falls in line with the national polls.