2008
Election Model: Predicted the Electoral
and Popular vote…but not the True Vote
The Final 2008
Election Model (EM) projection exactly matched the 365-173 electoral
vote and was close to matching the current popular vote share
(52.87-45.62%). The model projected
that Obama would receive 365.3 expected electoral votes with a 53.1- 44.9%
share. Obama currently leads by a
69.46-59.94m out of 131.37m recorded votes.
The EM projected that Obama would win by 75.9-64.2m out
of 143m total votes cast and capture 60% of undecided voters (the base case UVA
scenario). The 75% UVA scenario gave him 53.9% (or 55.0% of the two-party vote)
and 379.5 EV.
But the landslide was
denied . Obama did better than the EM projection and the recorded vote.
According to the 2004 U.S. Vote Census, 122.3m votes were recorded out of
125.7m cast; in 2000, 105.4m were recorded of 110.8m cast. Democrats
traditionally win 70-80% of the uncounted vote.
The 2008
Election Calculator (EC)
projected that Obama would win the True Vote with a 54.5-44.4% share
(78.3-63.8m).
The EC was run post-election based on 2008 National
Exit Poll vote shares, but with the returning voter mix changed from an impossible
46/37/4 Bush/Kerry/Other to a feasible mix based on a) the 2004 recorded vote
(implausible) and b) the 2004 unadjusted exit poll (plausible). Obama won the
implausible scenario (a) with 55.4% of the vote; he won the plausible scenario
(b) with 57.2%.
On Election Day, Obama led by 63.4-56.1m (52.3-46.3%)
with 121.2m votes recorded, so he has a 59.2-37.5% share of the late 10.16m
votes. Gore and Kerry also had late vote share margins that exceeded the
initial margin by 6-7%.
Obama’s expected EV was calculated as the product sum of
the state win probabilities and corresponding electoral vote.
In equation form: Expected EV =ĺ Win
Prob(i) * EV(i), for i= 1,51 states.
The state win probability is a function of the final 2-3
average polling spread and 3% margin of error (1200-1800 total sample). These
parameters are input to the Excel normal distribution probability function to
calculate the win probability. The expected 365.3 EV was close to the
365.8 average EV obtained in the Monte Carlo simulation. Obama won all
5000 election trials; his electoral vote win probability was 100%.
Right wing pundits are claiming that Obama’s 9.5 million
margin is not a mandate, unlike the Bush 3.0m “mandate” in 2004. But Obama won by more than that. The True
Vote will only be revealed if there is a real investigation by the courts,
Congress and/or the Media. Don’t hold your breath.
|
2004 |
True Vote |
|
|
|
2008 |
Calculated True Vote |
|
|
|
|
|
|
Cast |
Deaths |
Alive |
|
Turnout |
Voted |
Mix |
Obama |
McCain |
Other |
|
|
|
|
|
|
DNV |
29.9 |
20.8% |
57% |
41% |
2% |
|
Kerry |
66.9 |
3.2 |
63.7 |
|
95% |
60.6 |
42.2% |
91% |
8% |
1% |
|
Bush |
57.1 |
2.8 |
54.3 |
|
95% |
51.6 |
35.9% |
10% |
90% |
0% |
|
Other |
1.7 |
0.1 |
1.7 |
|
95% |
1.6 |
1.1% |
64% |
17% |
19% |
|
|
|
|
|
|
|
|
|
|
|
|
|
Total |
125.8 |
6.0 |
119.7 |
|
113.7 |
143.7 |
100% |
54.53% |
44.43% |
1.05% |
|
|
|
|
|
|
|
|
143.7 |
78.3 |
63.8 |
1.5 |
|
|
|
|
|
|
|
|
2-party |
55.1% |
44.9% |
|
The 2008 National Exit Poll – an impossible returning voter mix
|
Implied |
2004 |
Total |
Mix |
Obama |
McCain |
Other |
Obama |
McCain |
Other |
|
Vote |
DNV |
17.08 |
13% |
71% |
27% |
2% |
12.1 |
4.6 |
0.3 |
|
42.5% |
Kerry |
48.61 |
37% |
89% |
9% |
2% |
43.3 |
4.4 |
1.0 |
|
52.9% |
Bush |
60.43 |
46% |
17% |
82% |
1% |
10.3 |
49.6 |
0.6 |
|
4.6% |
Other |
5.25 |
4% |
66% |
24% |
10% |
3.5 |
1.3 |
0.5 |
|
114.3 |
Total |
131.37 |
100% |
52.62% |
45.52% |
1.86% |
69.13 |
59.80 |
2.44 |
The Post-election 2008 Election Calculator – two scenarios
The Election
Calculator used Final 2008 NEP vote shares but not the impossible
NEP return voter mix.
The assumptions were the following:
2004: 3.45 million uncounted (2.74% based on Census data), 4.8% mortality (table)
Turnout: 97% of living 2004 voters in 2008 (based on best estimates of 2008 votes cast and returning 2004 voters)
Uncounted votes in 2008: 3% of total votes cast (estimated)
The first scenario used
the recorded 2004 vote (Bush 50.73 - Kerry 48.27%) as the basis for the
returning voter mix.
Obama won by 16.8
million votes – a 7.3 million increase over his recorded margin.
The second scenario used
the unadjusted state exit poll aggregate (Kerry 52 - Bush 47%) as the basis for
the mix.
Obama won by 21.9
million - a 12.4 million increase over his recorded margin.
Scenario 1 (implausible): returning voter mix based on the recorded
2004 vote.
Obama wins by 16.81m: 75.01-58.20 (55.38-42.98%)
|
|
|
|
|
|
|
2008 |
|
Calculated Vote |
|
|
|
|
2004 |
Recorded |
Uncounted |
Cast |
Deaths |
Alive |
Turnout |
Voted |
Mix |
Obama |
McCain |
Other |
|
DNV |
|
|
|
|
|
|
19.42 |
14.34% |
71% |
27% |
2% |
|
Kerry |
59.03 |
2.58 |
61.61 |
2.96 |
58.65 |
96.9% |
56.85 |
42.0% |
89% |
9% |
2% |
|
Bush |
62.04 |
0.83 |
62.87 |
3.02 |
59.85 |
96.9% |
58.01 |
42.8% |
17% |
82% |
1% |
|
Other |
1.23 |
0.03 |
1.26 |
0.06 |
1.20 |
96.9% |
1.16 |
0.86% |
66% |
24% |
10% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total |
122.30 |
3.45 |
125.74 |
6.04 |
119.70 |
116.0 |
135.43 |
100% |
55.38% |
42.98% |
1.64% |
|
|
|
|
|
|
|
|
Cast |
135.43 |
75.01 |
58.20 |
2.22 |
|
|
|
|
|
|
|
|
|
Recorded |
52.87% |
45.62% |
1.51% |
|
|
|
|
|
|
|
|
|
131.37 |
69.46 |
59.94 |
1.98 |
Scenario 2
(plausible): returning voter mix based
on the unadjusted 2004 exit poll shares.
a) Obama wins by 21.86m: 77.52-55.66 (57.2-41.08%)
|
|
|
|
|
|
|
2008 |
|
Calculated Vote |
|
|
|
|
2004 |
Exit Poll |
Uncounted |
Cast |
Deaths |
Alive |
Turnout |
Voted |
Mix |
Obama |
McCain |
Other |
|
DNV |
|
|
|
|
|
|
19.42 |
14.34% |
71% |
27% |
2% |
|
Kerry |
63.59 |
1.79 |
65.38 |
3.14 |
62.25 |
96.9% |
60.33 |
44.5% |
89% |
9% |
2% |
|
Bush |
57.47 |
1.62 |
59.09 |
2.84 |
56.26 |
96.9% |
54.52 |
40.3% |
17% |
82% |
1% |
|
Other |
1.23 |
0.03 |
1.26 |
0.06 |
1.20 |
96.9% |
1.16 |
0.86% |
66% |
24% |
10% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total |
122.30 |
3.45 |
125.74 |
6.04 |
119.70 |
116.0 |
135.43 |
100% |
57.24% |
41.10% |
1.67% |
|
|
|
|
|
|
|
|
Cast |
135.43 |
77.52 |
55.66 |
2.26 |
|
|
|
|
|
|
|
|
|
Recorded |
52.87% |
45.62% |
1.51% |
|
|
|
|
|
|
|
|
|
131.37 |
69.46 |
59.94 |
1.98 |
2008
Model Obama
McCain
EM
53.1 - 44.9%, 365.3 EV; 60% UVA
(base case).
EM
53.9 - 44.1%, 379.3 EV; 75% UVA.
EC
54.5 - 44.5 %; feasible mix of returning 2004 voters; 12:22am NEP vote
shares.
Model Kerry Bush
EM 51.3 - 47.7% Kerry 337 EV;
75% UVA; unadjusted state aggregate exit poll:
52-47%.
EC 53.2 - 45.4 feasible mix
of returning 2000 voters; 12:22am NEP vote shares.
Exit Polls
WPE/IMS 52.0 - 47.0%
Kerry 337 EV; aggregate state unadjusted
NEP 51.4 - 47.7 (12:22am Voted in 2000 category)
Recorded 50.7 – 48.3%
Bush 286-252 EV
2004
Election Calculator
|
|
2004 Calculated |
||
|
|
Kerry |
Bush |
Other |
|
Cast |
66.94 |
57.07 |
1.74 |
|
Share |
53.23% |
45.39% |
1.38% |
|
EPoll (IMS) |
51.98% |
47.02% |
1.00% |
|
Switched |
5.32 |
-5.76 |
0.44 |
|
Sw. Rate |
8.0% |
-10.1% |
25.4% |
|
|
2004 Recorded |
|
|
|
Kerry |
Bush |
Other |
Total |
|
59.027 |
62.041 |
1.225 |
122.293 |
|
48.27% |
50.73% |
1.00% |
|
|
Uncounted |
|
|
|
|
2.588 |
0.794 |
0.069 |
3.450 |
|
75% |
23% |
2% |
2.74% |
2000
Election Calculator
|
|
2000
Calculated |
||
|
|
Gore |
Bush |
Other |
|
Cast |
55.287 |
51.431 |
4.081 |
|
Pct |
49.90% |
46.42% |
3.68% |
|
Exit Poll |
48.5% |
46.2% |
5.3% |
|
Switched |
0.25 |
-0.10 |
-0.14 |
|
Sw. Rate |
0.4% |
-0.2% |
-3.5% |
|
|
2000 Recorded |
|
|
|
Gore |
Bush |
Other |
Total |
|
51.004 |
50.456 |
3.957 |
105.417 |
|
48.38% |
47.86% |
3.75% |
|
|
Uncounted |
|
|
|
|
4.039 |
1.077 |
0.269 |
5.385 |
|
75% |
20% |
5% |
4.86% |