PICK A CARD, ANY CARD
Richard Hayes Phillips, Ph.D.
April 11, 2006
We have alleged since shortly after the 2004 presidential election that in
many counties in Ohio, thousands of votes were shifted from Kerry to Bush.
Comparison of the 2004 presidential results with other contests, on a
precinct by precinct basis, reveals anomalies that are simply not
explainable except by alteration of the vote count. In the forefront of
these is Warren County, in southwestern Ohio.
In Warren County, compared to 2000, the population increased by 14.75%,
voter registration by 29.66%, and ballots cast by 36.23%. Bush’s point
spread increased from 42.24% to 44.48%, and Bush’s victory margin
increased from 29,176 votes to 41,993 votes. Voter registration was
reportedly up by 79.0%, 38.3%, 32.4%, 31.0%, 29.7%, and 28.4% in six
townships that provided 68.75% of Bush’s margin of victory.
Countywide, Kerry received 2,427 fewer votes than C. Ellen Connally, an
underfunded African-American municipal judge from Cleveland running for
Chief Justice against Thomas J. Moyer, a well-financed Republican
incumbent, in a race that drew 1.2 million fewer votes statewide than the
presidential race. Precinct analysis shows that at least 2,557 persons
must have voted for both Connally and Bush. And for every one of the
22,081 persons who did not vote for Chief Justice but voted for Kerry,
there must have been another person who voted for both Connally and Bush.
Either that, or the official results are not true and correct.
Countywide, Bush received 4,606 more votes than Issue One, the
constitutional amendment prohibiting gay marriage, which passed
overwhelmingly with 61.71% of the vote statewide, compared to 50.81% for
Bush. (A vote for Issue One was a vote against gay marriage). Precinct
analysis shows that in six townships alone, at least 2,904 persons must
have voted for gay marriage and for George W. Bush. And for every person
who voted for Kerry and against gay marriage, there must have been another
person who voted for Bush and for gay marriage. Either that, or the
official results are not true and correct.
We believe that these numbers are fraudulent, in that at least 2,900
votes, and possibly 4,600 votes, have been shifted from Kerry to Bush,
resulting in a net loss of 5,800 votes to 9,200 votes for Kerry.
We have obtained photographs of the actual punch card ballots for 12 of
157 precincts in Warren County, and photographs of the poll books and
voter signature books for most of these 12 precincts. I recounted one
precinct, selecting one that exhibited both the “Connally anomaly” and the
“gay-friendly Republican” anomaly. Ballot by ballot, I recorded how each
person voted for president, for Chief Justice, and on Issue One.
This was not an easy task. In Ohio, the order in which the candidates’
names appear on the ballot must rotate from precinct to precinct. In
addition, one must examine the order in which the various races appear on
the precinct canvass records in order to determine which races appear in
which columns on the punch cards. In this particular precinct in Mason
City (149 MAS C WC AFS), the ballot rotation code was as follows:
Bush 8, Kerry 10
Connally 78, Moyer 79
Yes 124, No 125
We do not have photographs of every single ballot in this precinct, but
the data set is so nearly complete as to confirm that the hand count
matches the official results. In other words, the central tabulator was
working properly, and recording the punches correctly.
In the table below, Bush voters are on the left, and Kerry voters are on
the right. The first grouping shows those who voted on all three
questions – President, Chief Justice, and Issue One. For example, 244 of
those who voted for Moyer and for Issue One also voted for Bush. The
second grouping shows those who cast a vote either for Chief Justice or on
Issue One, but not both. For example, all 7 of those who voted for Moyer,
but did not vote on Issue One, also voted for Bush. The third grouping
shows those who voted for president but did not vote either for Chief
Justice or on Issue One. Bush got 19 of these voters and Kerry got 5.
Finally, there were 8 ballots for neither Bush nor Kerry.
HAND COUNT, WARREN COUNTY, MASON CITY, PRECINCT 149 MAS C WC AFS
BUSH VOTERS KERRY VOTERS
Yes No Yes No
Moyer (R) 244 62 Moyer 18 25
Connally (D) 109 42 Connally 22 76
Moyer 7 Moyer 0
Connally 1 Connally 3
Yes 71 Yes 16
No 19 No 28
Bush 19 Kerry 5
Punch Number 9 1
Punch Numbers 7 & 8 1
Bush got 313 (87.9%) of the Moyer voters
152 (60.1%) of the Connally voters
424 (88.3%) of the Yes voters
123 (48.8%) of the No voters
This is what the punch cards show. I struggle to believe it. One might
expect Bush to receive the votes of 87.9% of those who voted Republican
for Chief Justice, and 88.3% of those who voted against gay marriage. But
how could Bush have gotten 60.1% of the Ellen Connally voters, and 48.8%
of those who voted for gay marriage? Bush even got 42 (35.6%) of those
who voted for Ellen Connally and against Issue One.
I cannot help but notice the 19 ballots punched for Bush with no vote for
Chief Justice or on Issue One. Four of these punch cards contain only one
punch, for Bush. Six others contain only two punches: three for Bush and
Senator Voinovich, three for Bush and Issue One. This is consistent with
the possibility that the ballot box was stuffed with ten or more ballots
punched for Bush and for few, if any, other candidates. If this happened,
the voter signature book should reveal an equivalent number of forgeries.
And there are two ballots with odd-numbered punches in the presidential
column. Only the even-numbered spaces were valid for presidential
candidates. In this precinct, 8 was Bush, and 10 was Kerry. Yet one card
was punched at 7 and another at 9. The voting machines did not let the
voters do this. Board of Elections employees in Carroll, Harrison,
Muskingum, Seneca, and Wayne counties, all of which ran clean elections,
have confirmed that the ballots were covered by a “mask” or “shield” in
which holes were created for every possible voting position. It was not
possible to punch an odd number for president, unless the punch card was
outside the machine at the time. Also I notice that there were only 4
actual undervotes out of 775 ballots cast, a remarkably low percentage
(0.5%) for punch card ballots. This is consistent with the possibility
that undervotes were punched for Bush and fed back through the tabulators.
If this was done in a hurry, 7 and 9 could have been punched by mistake.
But neither of these possibilities, fraudulent though they be, can explain
the large anomalies in the vote totals. The explanation may lie in the
punches at the bottom of the cards. Each punch card ballot contains a
triangular array of punches outside the 228 numbered spaces that could
correspond to a candidate or a ballot issue. According to Board of
Elections employees in Harrison and Seneca counties, the punch code at the
bottom of the card matches the code on a “header card” which identifies
the precinct when the punch card ballots are run through the tabulator.
In Warren County we have photographs of punch card ballots for 12 of 157
precincts. The punch code at the bottom of the cards is the same for
every precinct. Near the bottom of every punch card appear these words:
TO BE FILLED IN BY ELECTION BOARD ONLY
Beneath this are the following words:
PRECINCT NO. ________________________
This line is always blank. Nowhere on the punch cards is the precinct
identified by punch code, and nowhere on the front of the punch cards is
the precinct identified in writing.
According to Dave Keeler, President of Dayton Legal Blank, Inc., which
prints ballots for 70 of 88 counties in Ohio, the precinct must be printed
on the back of each punch card ballot, by law. Board of Elections
employees in Muskingum and Seneca counties have confirmed that the
precincts were identified on the backs of the punch cards used in their
counties. We do not know if the precincts were identified on the backs of
the punch cards used in Warren County, because our photographers were not
allowed to touch the punch cards, and they never saw the back side of any
of them. Employees of the Warren County Board of Elections placed the
cards in stacks, front side up, on tables, spread them out eight at a time
to be photographed, and never, ever, turned them over. Thus I cannot
eliminate the possibility that ballots punched for Kerry were shifted to
other precincts where, due to ballot rotation, they would be tabulated as
votes for Bush. I shall explain the methodology as simply as possible.
The precinct canvass records list the precincts in a certain order. For
the first precinct on the list, the names of the candidates for each
office appear in alphabetical order. The positions shift one column to
the left as one moves down the list of precincts. There were five
candidates for President, and the rotation was as follows:
2 4 6 8 10
Precinct 1 Badnarik Bush Kerry Nader Peroutka
Precinct 2 Bush Kerry Nader Peroutka Badnarik
Precinct 3 Kerry Nader Peroutka Badnarik Bush
Precinct 4 Nader Peroutka Badnarik Bush Kerry
Precinct 5 Peroutka Badnarik Bush Kerry Nader
There were only two candidates for Chief Justice, so the rotation was as
Odd numbered precincts Connally Moyer
Even numbered precincts Moyer Connally
Ballot initiatives are not subject to ballot rotation, so the ballot
positions for Issue One were as follows:
All precincts Yes No
In Precinct 149 (MAS C WC AFS), Bush was in position 8, and Kerry was in
position 10. If, say, 30 ballots punched for Kerry were moved to Precinct
148, where Bush was in position 10, the votes would be counted for Bush.
The votes for Chief Justice would be reversed. In Precinct 149, Connally
got 101 (52.3%) of the Kerry voters, and Moyer got 43 (22.3%) of the Kerry
voters. Connally would be expected to get 16 of 30 Kerry voters, and
Moyer would be expected to get 7 of 30 Kerry voters. If the punch cards
were shifted to Precinct 148, Bush would get the 30 Kerry votes, Connally
would get the 7 Moyer votes, and Moyer would get the 16 Connally votes.
Kerry, instead of running 14 votes ahead of Connally, would run 7 votes
behind her, thus creating the “Connally anomaly.” The votes on Issue One
would remain unchanged. 129 (66.8%) of Kerry supporters voted “No” on
Issue One. One would expect Issue One to be opposed by 20 of 30 Kerry
supporters. If the ballots were shifted, these 20 gay-friendly Democrats
would become gay-friendly Republicans in the official vote count.
These 30 punch cards would not necessarily have gone to Precinct 148. The
same thing could be accomplished by shifting them to Precinct 153, or to
Precinct 3, or to Precinct 8, or to Precinct 13, or to any precinct where
Bush was in position 10 on the punch card ballots.
Conversely, if ballots punched for Kerry in Precinct 149 were moved to
Precinct 150, Kerry votes would go to Nader. These votes would disappear
into the black hole, because Nader had been disqualified, removed from the
ballot, and the tabulators were programmed to count his column as zero.
Again, the votes for Chief Justice would be reversed, and the votes on
Issue One would remain unchanged.
Such a systematic fraud could easily have been accomplished in Warren
County, given the peculiar circumstances of Election Night. The punch
card ballots collected from each precinct were driven to Lebanon, Ohio,
mainly along Interstate 71 and U. S. Highway 42, and brought to a tent,
equipped with metal detectors, on county property. It was decided that
the tent was too small, so the ballots were brought to an unauthorized
warehouse owned by the county, where they remained for four hours. During
this time it would have been easy to remove a predetermined number of
ballots punched for Kerry from each precinct, stack them up according to
his position on the ballot (2, 4, 6, 8, or 10), move the same number of
ballots to another precinct where that position would be counted as a vote
for Bush, and bind up the stack to signify that it was ready to run
through the central tabulator. It was only necessary to shift the same
number of ballots from precinct to precinct each time, so that the number
of ballots cast would match the number of names in the signature book.
After that, the ballots were brought to the county administrative
building, which was locked down by county officials, who prohibited all
independent observers from watching the vote count. Citing “homeland
security” concerns, the county had two pickup trucks out front, police in
the building, and a bomb-sniffing dog. No one would know if the tabulator
was able to identify the punch cards by precinct, or if the tabulator had
to be commanded manually, because no independent observers were watching.
All of this is completely consistent with the official election results
for Warren County. It would explain how John Kerry got 2,427 (8.52%)
fewer votes than Ellen Connally, while running 32.18% ahead of her
statewide. It would explain how George W. Bush got 4,606 (7.26%) more
votes than Issue One, while running 14.10% behind Issue One statewide.
And it would explain how, in Precinct 149 (MAS C WC AFS), Bush got 60.1%
of the Ellen Connally voters, and 48.8% of the gay marriage supporters.
The statistical and circumstantial evidence is powerful. There is a prima
facie case that a crime has been committed. It should be up to the county
officials to provide convincing exculpatory evidence. I want them to pick
a card, any card, and show me the back of it, so that I can see for myself
if it is identified by precinct. Then I want them to pull the cards for
precincts of my choosing, and to deal the cards face down, so that I can
see for myself if all the cards belong to the same deck.