2008 Landslide Denied: Uncounted Votes and the Final
National Exit Poll
Dec. 20, 2008
The latest election
results show Obama leading by 9.5m votes (69.46-59.94m), a 52.87-45.62%
share of 131.37m. Obama must have done better than the 2008 Final
National Exit Poll (NEP) which is always forced to match the recorded vote.
He also did better than the 2008 Election Model
projection. The 2008
Election Calculator indicates that his vote share was 55-57%.
The tremendous GOTV and new registration effort has not resulted in the
expected increase in the recorded vote from 2004. There was a 17m net increase
from 2000 to 2004 (105m to 122m) compared to the 8.8m increase from 2004. The
media should be asking why.
On Election Day, Obama led
by 63.4-56.1m (52.3-46.3%) with 121.2m votes recorded. Since then, he has a
59.2-37.5% share of the 9.9m late votes recorded. Gore and Kerry also won the
late vote by much more than the initial margin.
Historically, about 75%
of uncounted votes (spoiled, lost, provisional and absentee) are Democratic.
But we will have to wait until the Census Bureau publishes its 2008 vote survey
to know how many votes were cast.
In order to match the
vote count, the Final NEP adjusts demographic category weights and vote shares.
The Voted in 2004 category indicates an IMPOSSIBLE 46 Bush/37% Kerry returning
voter split of the 2008 electorate! Analysts would like to know what the split
was in the earlier NEP update - before the Final was matched to the vote. We
have the preliminary numbers from 2004 and the 2006 midterms.
Changes to the prior
election mix of returning voters from the Preliminary to the Final NEP always
move in favor of GOP. The goal is to MATCH THE RECORDED VOTE COUNT. This
election was no different. How does the official 50.7-48.3% Bush recorded margin
in 2004 square with the 46/37% returning voter mix in the Final 2008 NEP? The
anomaly is even more ridiculous since Kerry won the 2004 True Vote. The number
of returning (motivated) Kerry voters in 2008 had to exceed the number of
returning (unmotivated) Bush voters by a similar amount.
2004 NEP (Voted in 2000)
The returning Bush/Gore
voter mix was 41/39% at 12:22am; it was changed to an impossible 43/37% in the
Final. Election stolen.
2006 NEP (Voted in 2004)
The returning Bush/Kerry
voter mix was 47/45% at 7pm; it was changed to an implausible 49/43% in the
2008 Final National Exit Poll.
Landslide denied.
2008 NEP (Voted in 2004)
The preliminary NEP
Bush/Kerry voter mix is not yet available; it is an impossible 46/37%
in the Final. Landslide denied.
Uncounted votes are an important
component in the discrepancy between the polls (pre-election and exit) and the
recorded vote. The Democrats ALWAYS do better in the polls than in the vote
count. In 2004, Bush won the "official" vote by 50.7-48.3% but Kerry
won the preliminary, adjusted 2004
Preliminary 12:22am National Exit Poll (13047 respondents) by 51-48%. The
Final NEP is always forced to match the recorded vote without consideration of
uncounted and/or switched votes.
In the 2000 election, 105.4m votes
were recorded out of 110.8m cast. The 5.4m uncounted ballots were a combination
of provisional, spoiled, lost and absentees. Therefore, we estimate that Al
Gore won by at least 3m votes, not by the 540k recorded.
In 2004, 122.3m votes
were recorded out of 125.7m cast (3.4m were uncounted). The 1.5m net votes to
Kerry cut the Bush margin in half. And that is before vote suppression,
stuffing and switching. The 2004 Election
Calculator indicates that Kerry won by 67-57m, based on a feasible (and
plausible) number of returning Gore and Bush 2000 voters (4.9% mortality, 95%
turnout), 3.45m uncounted votes and 12:22am NEP vote shares.
Once again, we need the
unadjusted precinct exit poll data, not the Final National and State exit polls
that are adjusted to match the recorded vote count. The media never releases that
information, claiming the need for voter confidentiality. But analysts are not
interested in the names of the respondents, just the pristine precinct exit
poll data.
An analysis of Uncounted Votes and Exit
Poll Discrepancies from 1988-2004 reveals some very interesting
information. In the 1988-2000 elections, the unadjusted state exit poll
aggregate vote share was within 1% of the recorded vote after it was adjusted
to include uncounted votes. But in 2004 adding the uncounted votes was not
enough to make up the difference. HAVA (Help America Vote Act) was passed
during the first Bush term and many new touch screen voting machines were
installed nationwide.
The vast majority (90%) of vote-switching
incidents at touch screens are from the Democrat to the Republican. We must
assume that vote-switching also occurred on the central tabulators which sum
the precinct vote totals for all voting machines, including optical scanners.
The touch screens have no paper trail, so a vote recount is impossible. But
this analysis will focus on uncounted votes.
In Jan. 2005, exit pollsters
Edison-Mitofsky report gave the unadjusted state exit poll deviations (WPE)
from the recorded vote.
The state exit polls indicated that
Kerry won by 52-47%.
The 2004 Election
Calculator model indicated that Kerry won the True Vote by 53.2-45.8%.
NEP 12:22am vote shares
were assumed with feasible 2000 voter weightings.
In 2006, the pre-election Generic
polls indicated that the Democrats would win by 56-42%.
The 7pm National Exit Poll had the
Democrats winning by 55-43%.
The Final NEP was forced to match
the 52-46% recorded vote count.
2008 Election Model
and 2008
Election Calculator
Election Model (Final
projection)
Scenario
1: Obama won 60% of the undecided votes (base case)
He won by 53.1-44.9% with
an expected 365.3 electoral votes.
Scenario 2: Obama won 75% of the
undecided votes
He won by 53.8-44.2% with an
expected 379.5 electoral votes.
Final 2008 NEP (Voted 2004)
Obama won by 52.6-45.5%. The
returning voter mix was impossible.
Election Calculator (post-election)
Scenario 1: Final 2008 NEP vote
shares with the returning voter mix based on the recorded 2004 vote.
Obama won by 55.7-42.7% (75.4-57.8m).
Scenario 2: Final 2008 NEP vote
shares with the returning voter mix based on the unadjusted 2004 state exit
polls (WPE/IMS method)
Obama won the True Vote by
57.5-40.8% (77.9-55.3m).
Actual NEP: 2004 returning voter mix forced to
match the 2008 recorded vote
Obama margin: 9.3m
|
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 |
Note: Of those who did not vote in
2004, Obama won 68% of first-timers and 88% of others who voted prior to 2004.
Scenario 1: Adjusted
2004 returning voter mix based on 2004 recorded vote and
2004 voters: 3.45
million uncounted (2.74%), 4.8% mortality, 95% turnout; 3.0% uncounted in 2008
a) Obama wins by 17.6m:
75.4-57.8 (55.7-42.7%) assuming 4.0m uncounted votes (3.0% of 135.4m cast).
b) Obama wins by 15.8m:
72.5-56.7 (55.2-43.1%) assuming no uncounted votes.
|
Recorded |
2004 |
Total |
Mix |
Obama |
McCain |
Other |
Obama |
McCain |
Other |
|
Vote |
DNV |
21.71 |
16.0% |
71% |
27% |
2% |
15.4 |
5.9 |
0.4 |
|
48.27% |
Kerry |
55.72 |
41.1% |
89% |
9% |
2% |
49.6 |
5.0 |
1.1 |
|
50.73% |
Bush |
56.86 |
42.0% |
17% |
82% |
1% |
9.7 |
46.6 |
0.6 |
|
1.00% |
Other |
1.14 |
0.8% |
66% |
24% |
10% |
0.8 |
0.3 |
0.1 |
|
|
Total |
135.43 |
100.0% |
55.69% |
42.66% |
1.65% |
75.43 |
57.77 |
2.23 |
Scenario 2: Adjusted 2004 returning voter mix based on 2004 unadjusted
exit poll and
2004 voters: 3.45 million uncounted (2.74%), 4.8% mortality, 95%
turnout; 3.0% uncounted in 2008
a) Obama wins by 22.6m: 77.9-55.3 (57.5-40.8%)
assuming 4.0m uncounted votes.
b) Obama wins by 20.8m: 75.0-54.2 (57.1-41.3%)
assuming no uncounted votes.
|
2004 Unadj. |
2004 |
Total |
Mix |
Obama |
McCain |
Other |
Obama |
McCain |
Other |
|
Exit Poll |
DNV |
21.71 |
16.0% |
71% |
27% |
2% |
15.4 |
5.9 |
0.43 |
|
52.0% |
Kerry |
59.13 |
43.7% |
89% |
9% |
2% |
52.6 |
5.3 |
1.18 |
|
47.0% |
Bush |
53.44 |
39.5% |
17% |
82% |
1% |
9.1 |
43.8 |
0.53 |
|
1.0% |
Other |
1.14 |
0.8% |
66% |
24% |
10% |
0.8 |
0.3 |
0.11 |
|
|
Total |
135.43 |
100% |
57.51% |
40.82% |
1.67% |
77.88 |
55.28 |
2.27 |
Returning voter mix based on 2000 Recorded vote
|
|
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% |
|
|
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 Recorded |
|
|
|
|
|
|
2004 Calculated |
|
|
||
|
Voted |
Recorded |
Uncounted |
Cast |
Deaths |
Alive |
|
Turnout |
Voted |
Weight |
Kerry |
Bush |
Other |
|
|
|
|
|
|
|
|
DNV |
25.62 |
20.4% |
57% |
41% |
2% |
|
Gore |
51.00 |
4.04 |
55.04 |
2.72 |
52.32 |
|
95% |
49.70 |
39.5% |
91% |
8% |
1% |
|
Bush |
50.46 |
1.08 |
51.53 |
2.48 |
49.06 |
|
95% |
46.60 |
37.1% |
10% |
90% |
0% |
|
Nader/other |
3.96 |
0.27 |
4.23 |
0.21 |
4.02 |
|
95% |
3.82 |
3.0% |
64% |
17% |
19% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total |
105.42 |
5.38 |
110.80 |
5.41 |
105.39 |
|
100.13 |
125.74 |
100% |
53.23% |
45.39% |
1.38% |
|
|
|
|
|
|
|
|
|
|
125.74 |
66.94 |
57.07 |
1.74 |
|
|
|
|
|
|
|
|
|
|
|
|
||
|
|
|
Sensitivity Analysis |
|
|
|
|
Sensitivity Analysis |
|
||||
|
|
|
Kerry National Vote |
|
|
|
|
Kerry National Vote |
|
||||
|
Gore share of |
|
|
|
|
|
|
Kerry share of |
|
|
|
|
|
|
Uncounted in 2000 |
Gore Voter Turnout |
|
|
|
Gore voters |
Share of New voters (DNV in 2000) |
||||||
|
53.2% |
91.0% |
93.0% |
95.0% |
97.0% |
99.0% |
|
53.2% |
53.0% |
55.0% |
57.0% |
59.0% |
61.0% |
|
95.0% |
53.3% |
53.6% |
53.9% |
54.2% |
54.5% |
|
95.0% |
54.0% |
54.4% |
54.8% |
55.2% |
55.6% |
|
85.0% |
53.0% |
53.3% |
53.6% |
53.9% |
54.1% |
|
93.0% |
53.2% |
53.6% |
54.0% |
54.4% |
54.8% |
|
75.0% |
52.7% |
52.9% |
53.2% |
53.5% |
53.8% |
|
91.0% |
52.4% |
52.8% |
53.2% |
53.6% |
54.0% |
|
65.0% |
52.3% |
52.6% |
52.9% |
53.2% |
53.5% |
|
89.0% |
51.6% |
52.0% |
52.4% |
52.8% |
53.3% |
|
55.0% |
52.0% |
52.3% |
52.6% |
52.9% |
53.1% |
|
87.0% |
50.8% |
51.2% |
51.7% |
52.1% |
52.5% |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Kerry Margin (millions) |
|
|
|
|
Kerry Margin (millions) |
|
||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9.87 |
91.0% |
93.0% |
95.0% |
97.0% |
99.0% |
|
9.87 |
53.0% |
55.0% |
57.0% |
59.0% |
61.0% |
|
95.0% |
10.1 |
10.8 |
11.5 |
12.3 |
13.0 |
|
95.0% |
11.8 |
12.8 |
13.8 |
14.9 |
15.9 |
|
85.0% |
9.2 |
10.0 |
10.7 |
11.4 |
12.2 |
|
93.0% |
9.8 |
10.8 |
11.9 |
12.9 |
13.9 |
|
75.0% |
8.4 |
9.1 |
9.9 |
10.6 |
11.3 |
|
91.0% |
7.8 |
8.8 |
9.9 |
10.9 |
11.9 |
|
65.0% |
7.6 |
8.3 |
9.0 |
9.8 |
10.5 |
|
89.0% |
5.8 |
6.9 |
7.9 |
8.9 |
9.9 |
|
55.0% |
6.8 |
7.5 |
8.2 |
8.9 |
9.6 |
|
87.0% |
3.8 |
4.9 |
5.9 |
6.9 |
7.9 |