MONTE CARLO SIMULATION
HOUSE ELECTION FORECAST
61 GOP House seats
Democrats need to win 15 to gain control.
Sensitivity Analysis
--------------------
Base case: 60% of undecided voters (UVA) break for Democratic challengers.
UVA 50 55 60 67 75
N Probability (Dems win at least N races)
41 80 97 99.9 100 100
43 30 71 94 100 100
45 2 15 48 89 100
Probability that Democrats win at least N seats:
N 41 42 43 44 45 46 47 48
Prob 99.9 98.8 94 79 48 17 4 0.1
Here's a graph:
http://www.geocities.com/electionmodel/Election2006_22243_image001.png
New House make-up:
Dems 243 244 245 246 247 248 249 250
GOP 192 191 190 189 188 187 186 185
___________________________________________________________________
ELECTION FRAUD SCENARIO ANALYSIS
---------------------------------
Current Projections (Pre-hack, assume ZERO fraud):
Dems win 44 of 61 GOP seats and capture the House 246-189.
Average Vote share: Dem 52% - GOP 48%
Fraud Scenario Assumptions
(required for GOP to Win House):
1. 3% of votes are not counted (75% Democratic)
2. 11% of Democratic votes are switched to GOP
Post-hack (MASSIVE FRAUD):
31 elections are switched from Democrats to GOP
The Democrats win 13 seats, 2 shy of the 15 required for a majority.
The GOP wins 47 seats and retain the House.
Vote share: GOP 53.9% - Dem 46.1%
This is a graph of the relationship between the percentage of switched votes
and the number of stolen House elections:
http://www.geocities.com/electionmodel/Election2006Final1105a_21869_image001.png
A graph of district vote shares - before and after hacking:
http://www.geocities.com/electionmodel/Election2006Final1105a_16256_image001.png
Projections based on latest polls (60% UVA to Dems)
---------------------------------------------------
Pre-hack Post-hack
District Oct.26 Dem GOP Nov.3 Dem GOP Dem GOP Prob Dem GOP Switch
Code Average 46.9 46.5 Latest 47.0 44.6 52.0 48.0 98.0 46.1 53.9 31
1 AZ 1 10/26 46 48 10/26 46 48 49.6 50.4 34 43.5 56.5
2 AZ 5 10/15 45 48 10/15 48 46 51.6 48.4 95 45.6 54.4 yes
3 AZ 8 10/2 45 37 10/29 53 41 56.6 43.4 100 50.9 49.1
4 CA 4 10/26 46 49 10/26 39 39 52.2 47.8 99 46.2 53.8 yes
5 CA 11 10/26 48 46 10/26 48 46 51.6 48.4 95 45.6 54.4 yes
6 CA 50 9/12 40 54 9/12 40 54 43.6 56.4 0 37.2 62.8
7 CO 4 10/26 48 45 11/3 43 44 50.8 49.2 80 44.7 55.3 yes
8 CO 5 10/18 38 51 10/27 40 47 47.8 52.2 1 41.6 58.4
9 CO 7 10/26 51 46 10/31 54 38 58.8 41.2 100 53.3 46.7
10 CT 2 10/26 51 45 10/29 48 47 51.0 49.0 85 45.0 55.0 yes
11 CT 4 10/26 42 52 10/30 51 44 54.0 46.0 100 48.2 51.8 yes
12 CT 5 10/26 51 43 10/29 46 43 52.6 47.4 100 46.7 53.3 yes
13 FL 13 10/26 49 47 10/26 49 47 51.4 48.6 93 45.4 54.6 yes
14 FL 16 10/13 48 41 10/13 48 41 54.6 45.4 100 48.8 51.2 yes
15 FL 22 10/26 50 48 10/26 50 48 51.2 48.8 89 45.2 54.8 yes
16 FL 24 10/24 43 45 10/24 43 45 50.2 49.8 58 44.1 55.9 yes
17 ID 1 10/26 37 39 11/1 38 34 54.8 45.2 100 49.0 51.0 yes
18 IL 6 10/26 48 47 10/29 54 40 57.6 42.4 100 52.0 48.0
19 IL 10 10/26 48 46 10/26 48 46 51.6 48.4 95 45.6 54.4 yes
20 IL 14 10/10 42 52 10/10 42 52 45.6 54.4 0 39.3 60.7
21 IL 19 10/10 36 53 10/10 36 53 42.6 57.4 0 36.2 63.8
22 IN 2 10/26 48 45 10/31 52 43 55.0 45.0 100 49.2 50.8 yes
23 IN 8 10/26 53 43 10/26 53 43 55.4 44.6 100 49.7 50.3
24 IN 9 10/26 51 43 10/26 51 43 54.6 45.4 100 48.8 51.2 yes
25 IA 1 10/8 48 50 10/29 49 42 54.4 45.6 100 48.6 51.4 yes
26 IA 2 10/26 48 47 10/26 48 47 51.0 49.0 85 45.0 55.0 yes
27 KY 3 10/26 52 46 11/1 52 44 54.4 45.6 100 48.6 51.4 yes
28 KY 4 10/26 50 46 10/29 45 42 52.8 47.2 100 46.9 53.1 yes
29 MN 1 10/26 47 50 10/26 47 50 48.8 51.2 11 42.6 57.4
30 MN 2 10/15 42 50 10/15 42 50 46.8 53.2 0 40.5 59.5
31 MN 6 10/26 47 48 10/26 42 52 45.6 54.4 0 39.3 60.7
32 NV 3 10/26 44 51 10/30 39 46 48.0 52.0 2 41.8 58.2
33 NH 1 10/8 37 49 11/2 42 47 48.6 51.4 7 42.4 57.6
34 NH 2 10/26 50 47 11/2 45 37 55.8 44.2 100 50.1 49.9
35 NJ 7 10/26 43 46 10/26 43 46 49.6 50.4 34 43.5 56.5
36 NM 1 10/19 45 42 10/29 53 44 54.8 45.2 100 49.0 51.0 yes
37 NY 3 10/10 44 51 10/10 44 51 47.0 53.0 0 40.7 59.3
38 NY 19 10/26 49 47 10/26 49 47 51.4 48.6 93 45.4 54.6 yes
39 NY 20 10/26 53 42 10/26 53 42 56.0 44.0 100 50.3 49.7
40 NY 24 10/10 53 42 10/10 53 42 56.0 44.0 100 50.3 49.7
41 NY 26 10/18 46 49 10/26 53 44 54.8 45.2 100 49.0 51.0 yes
42 NY 29 10/26 43 42 10/26 53 42 56.0 44.0 100 50.3 49.7
43 NC 8 10/26 48 44 10/26 48 44 52.8 47.2 100 46.9 53.1 yes
44 NC 11 10/26 53 44 10/26 48 43 53.4 46.6 100 47.5 52.5 yes
45 OH 1 10/26 46 48 10/26 48 46 51.6 48.4 95 45.6 54.4 yes
46 OH 2 10/17 46 51 10/31 48 45 52.2 47.8 99 46.2 53.8 yes
47 OH 6 10/18 64 32 10/18 64 32 66.4 33.6 100 61.7 38.3
48 OH 15 10/10 53 41 10/10 53 41 56.6 43.4 100 50.9 49.1
49 OH 18 10/11 48 41 10/27 53 33 61.4 38.6 100 56.2 43.8
50 OK 5 10/10 33 62 10/10 33 62 36.0 64.0 0 29.4 70.6
51 PA 4 10/26 47 51 10/26 48 46 51.6 48.4 95 45.6 54.4 yes
52 PA 6 10/26 51 46 10/29 49 44 53.2 46.8 100 47.3 52.7 yes
53 PA 7 10/10 52 44 10/24 50 43 54.2 45.8 100 48.4 51.6 yes
54 PA 8 10/26 50 47 10/30 46 41 53.8 46.2 100 47.9 52.1 yes
55 PA 10 10/23 48 39 10/26 47 38 56.0 44.0 100 50.3 49.7
56 TX 22 na na na 10/25 36 28 57.6 42.4 100 52.0 48.0
57 VA 2 10/26 50 45 10/29 43 51 46.6 53.4 0 40.3 59.7
58 VA 5 10/10 40 56 10/10 40 56 42.4 57.6 0 36.0 64.0
59 VA 10 10/10 42 47 10/10 42 47 48.6 51.4 7 42.4 57.6
60 WA 8 10/26 49 47 10/30 45 51 47.4 52.6 0 41.2 58.8
61 WI 8 10/26 51 45 10/26 51 45 53.4 46.6 100 47.5 52.5 yes
_______________________________________________________________________
ELECTION FRAUD MODEL
Purpose:
Determine the level of fraud required to reverse the true vote
(assumed equal to the final poll).
Application to Final House polls:
Determine the level of fraud necessary in each district in order for the GOP to
steal the election. The analysis is based on the final pre-election polls, the
expected percentage split of uncounted votes, and the percentage undecided
voter split.
There are three primary categories:
1)Voter disenfranchisement
2)Uncounted votes (spoiled, lost, discarded, etc.)
3)Switched votes at the voting machine and/or central tabulator.
Although it's a major fraud component, disenfranchisement is not
factored into the model. True vote shares are matched to the final pre-election
polls. Approximately 3% of total votes cast are never counted.
The majority of uncounted votes are in minority districts.
Example:
a) Democrats lead 50-47% in the final pre-election poll.
b) 60% of undecided votes break for the Democratic challenger.
c) 4% of Democratic votes are switched to the GOP.
Note: The 2.36% "breakeven" vote switch is the calculated minimum
switch necessary for the election to end up a TIE.
ASSUMPTIONS:
Poll MoE MoE 3%
Dem Poll DP 50
Rep Poll RP 47
Undecided Vote UVA 60% to Dem
Spoiled Votes VSP 3% (assume 75% are Dem)
Switched Votes VSW 4% from Dem to Rep
Calculate: Dem Rep
PRE-HACK 51.80 48.20 (after UVA adj.)
Less:Spoiled 2.25 0.99
Adj. Vote share 49.55 47.21
2-Party Vote 51.21 48.79
Switched -2.05 2.05
POST-HACK 49.16 50.84
Breakeven Switch 2.36% (for 50/50 result)
Prob. of Discrepancy: 4.2% (2-party pre-hack less post-hack)
Sensitivity analysis:
Switched Votes 0% 1% 2% 3% 4% 5% 6% 7%
Probability 35% 24% 15% 8% 4.2% 2% 0.8% 0.3%
_____________________________________________________________________________