AB
1
 2012 Presidential True Vote and Election Fraud Simulation Model
2
3
Richard Charnin
4
http://richardcharnin.com/TrueVoteModels.htm
5
11/5/2012
6
7
8
The 2012 Presidential True Vote and Election Fraud Simulation Model (TVM) is a combination of 1) the pre-election Monte Carlo Simulation Election Model, based on the latest state polling and 2) a post-election True Vote Model, based on a feasible estimate of new and returning 2008 voters and corresponding estimated vote shares. The model will be updated periodically for the latest state and national polls.
9
10
The model projections assume the election is held on the latest poll date.
11
Obama needs a True Vote of at least 55% in order to overcome the historic 4-5% red-shift (the fraud factor).
12
13
Pre-election Polls and the”Horse Race”
14
15
The Real Clear Politics website archives the latest polls on a daily basis. The model uses 2008 recorded vote shares for states that have not yet been polled and the outcome is a near certainty...
16
http://www.realclearpolitics.com/epolls/latest_polls/elections/
17
18
Virtually all of the current national polls are of Registered Voters (RV). An exception is Rasmussen, a GOP pollster, whose daily tracking poll is a Likely Voter (LV) subset of the RV.
19
20
One month before the election, pollsters convert from the full RV sample to the LV subsample using the "Likely Voter Cutoff Model". The pollsters expect there will be fraud, so they prepare for it. The RV polls are transformed to LV subsets in order to promote an artificial “horse race”. The pundits want to predict the recorded vote. The closer they are, the better they look. But they never mention that it’s the fraud factor that gets them close. LV polls are usually good predictors of the fraudulent recorded vote.
21
22
The Likely Voter Cutoff Model effectively understates the potential turnout of millions of new Democratic voters who do not pass the LVCM. The pollsters and the pundits invariably claim that the race is "tightening" and LV polls are better predictors. And it's true: LV polls are better predictors of the fraudulent recorded vote.
23
24
25
1988-2008 Exit Poll Discrepancies
26
27
Based on the historic record, Obama needs at least 55% to overcome systemic built-in fraud. In 2008 Obama had a 58% share in the weighted average of the unadjusted STATE exit polls (76,000 respondents). He had 61% in the unadjusted National Exit Poll (17,836 respondents).
28
29
But for the same 17,836 respondents, the FINAL National Exit poll was forced to match the 52.9% recorded share. Obama’s 18% state aggregate unadjusted exit poll margin was reduced to 7.5% But that’s nothing; it is standard operating procedure on the part of exit pollsters IN EVERY ELECTION to ADJUST the unadjusted exit polls to match the the recorded vote THE ASSUMPTION IS ALWAYS THE SAME: ZERO FRAUD. IT'S A JOKE.
30
31
Election fraud reduced the average 1988-2008 Democratic presidential share by 4%. They won the unadjusted exit polls by 52-42% compared to the 48-46% recorded vote..
32
33
Democratic True Vote = Recorded Vote + Fraud Factor
34
35
Final exit polls are always forced to match the recorded votes that were predicted by the final pre-election LV polls.
36
37
The True Vote Model
38
39
The TVM has two options for estimating returning voters:
40
1- The True Vote option (default) assumes that 2008 voters return in the same proportion as the unadjusted exit poll aggregate (Obama by 58-40.5%).
41
Assuming they are proportional to the 2008 unadjusted state exit polls, Obama wins by 54.3-45.7% with 355 EV (99.99% win probability).
42
43
2- The Recorded Vote (Fraud) option assumes that voters return in proportion to the 2008 recorded vote (won by Obama 52.9-45.6%).
44
Assuming returning 2008 voters are proportional to the 2008 recorded vote, Obama wins the TVM by 50.4-49.6% with 284 EV (65% win probability).
45
46
The user can enter estimated turnout rates of living Obama and McCain voters.
47
A 1.25% annual voter mortality rate is assumed.
48
49
The TVM can be used to project a given state as well as the national vote. Turnout rates and vote shares used in projecting the national vote are applied to each state in order to derive the national aggregate.
50
51
Sensitivity analysis
52
53
It is important in election modeling to view the effects of changes in input parameters. The TVM displays the effects of turnout rates and share of returning voters. Three tables are generated for nine scenario combinations of a) Obama and McCain turnout rates and b) shares of returning Obama and McCain voters. The tables provide resulting vote shares, margins and popular vote win probabilities.
54
55
Monte Carlo Simulation: 500 election trials
56
57
There are two options for running the simulation model. Both should be used and the results compared. The default option is to use the TVM projected state vote shares. The second is to use the projected state pre-election polls.
58
59
The simulation consists of 500 election trials. The electoral vote win probability is percentage of winning election trials.
60
61
The projected state vote share is the sum of the poll share and the undecided voter allocation (UVA). The model uses state vote share projections as input to the Normal Distribution function to determine the state win probability.
62
63
In each election trial, a random number (RND) between 0 and 1 is generated for each state and compared to Obama’s state win probability. If RND is greater than the win probability, the Republican wins the state. If RND is less than the win probability, Obama wins the state. The winner of the election trial is the candidate who has at least 270 electoral votes. The process is repeated in 500 election trials.
64
65
2008 State Exit Poll and recorded vote data is displayed in the ‘2008‘ worksheet. The latest state polls are listed in the ‘Polls” worksheet which will be used for trend analysis. The data is displayed graphically in the ‘PollChart’ worksheet. A histogram of the Monte Carlo Simulation (500 trials) is displayed in the ‘ObamaEVChart’ worksheet.
66
67
Electoral Votes and Win Probabilities
68
69
The Electoral Vote is calculated in three ways.
70
71
1. The snapshot EV is a simple summation of state electoral votes. It could be misleading; there could be several close elections which go one way.
72
2. The Theoretical EV is the product sum of state electoral votes and win probabilities. A simulation or meta-analysis is not required to calculate the expected EV.
73
3. The Mean EV is the average electoral vote in the 500 simulated elections.
74
75
The Mean EV will be close to the Theoretical EV, illustrating the Law of Large Numbers. The snapshot EV will likely differ slightly from the Theoretical EV, depending on the number of state election projections that fall within the margin of error.
76
77
Obaama’s electoral vote win probability is the percentage of 500 simulated election trials that he won.
78
79
The national popular vote win probability is calculated as the state win probabilities: using the normal distribution. based on the final national share and the margin of error
80
81
The Fraud Factor
82
83
The combination of True Vote Model and state poll-based Monte Carlo Simulation enables the analyst to determine if the electoral and popular vote share estimates are plausible. The aggregate state poll shares can be compared to the default TVM (no fraud).
84
85
The TVM can be forced to match the aggregate poll projection by…
86
- Adjusting the vote shares (enter the incremental adjustment in cell C27). A red flag would be raised if the match required that Obama capture 85% of returning Obama voters while Romney gets 95% of returning McCain voters (a 10% net defection).
87
88
- Adjusting 2008 voter turnout in 2012 to force a match to the aggregate projected poll shares. For example, if McCain voter turnout is required to be 10-15% higher than Obama’s, that would also raise a red flag.
89
90
- Setting the recorded vote option (enter 1 in cell B18). If the resulting TVM vote shares indicate a close election and close to the projected poll shares, it would suggest that Romney has a good chance of winning.
91
92
Check the resulting electoral vote calculations and corresponding win probabilities.
93
94
95
96
97
98
99