George W. Bush’s vote
tallies, especially in the key state of Florida, are so statistically
stunning that they border on the unbelievable.
While it’s extraordinary for a candidate to get a
vote total that exceeds his party’s registration in any voting
jurisdiction – because of non-voters – Bush racked up more votes than
registered Republicans in 47 out of 67 counties in Florida. In 15 of
those counties, his vote total more than doubled the number of
registered Republicans and in four counties, Bush more than tripled the
Statewide, Bush earned about 20,000 more votes than
By comparison, in 2000, Bush’s Florida total
represented about 85 percent of the total number of registered
Republicans, about 2.9 million votes compared with 3.4 million
Bush achieved these totals although exit polls
showed him winning only about 14 percent of the Democratic vote
statewide – statistically the same as in 2000 when he won 13 percent of
the Democratic vote – and losing Florida’s independent voters to Kerry
by a 57 percent to 41 percent margin. In 2000, Gore won the independent
vote by a much narrower margin of 47 to 46 percent.
[For details on the Florida turnout in 2000, see
details on the 2004 Florida turnout, see
Exit Poll Discrepancies
Similar surprising jumps in Bush’s vote tallies
across the country – especially when matched against national exits
polls showing Kerry winning by 51 percent to 48 percent – have fed
suspicion among rank-and-file Democrats that the Bush campaign rigged
the vote, possibly through systematic computer hacking.
Republican pollster Dick Morris said the Election
Night pattern of mistaken exit polls favoring Kerry in six battleground
states – Florida, Ohio, New Mexico, Colorado, Nevada and Iowa – was
“Exit polls are almost never wrong,” Morris wrote.
“So reliable are the surveys that actually tap voters as they leave the
polling places that they are used as guides to the relative honesty of
elections in Third World countries. … To screw up one exit poll is
unheard of. To miss six of them is incredible. It boggles the
imagination how pollsters could be that incompetent and invites
speculation that more than honest error was at play here.”
But instead of following his logic that the
discrepancy suggested vote tampering – as it would in Latin America,
Africa or Eastern Europe – Morris postulated a bizarre conspiracy theory
that the exit polls were part of a scheme to have the networks call the
election for Kerry and thus discourage Bush voters on the West Coast. Of
course, none of the networks did call any of the six states for Kerry,
making Morris’s conspiracy theory nonsensical. Nevertheless, some
Democrats have agreed with Morris's bottom-line recommendation that the
whole matter deserves “more scrutiny and investigation.” [The
Hill, Nov. 8, 2004]
Democratic doubts about the Nov. 2 election have deepened with
anecdotal evidence of voters reporting that they tried to cast votes for
Kerry but touch-screen voting machines came up registering their votes
In Ohio, election officials said an error with an electronic voting
system in Franklin County gave Bush 3,893 extra votes in suburban
Columbus, more than 1,000 percent more than he actually got.
Yet, without a nationwide investigation, it’s impossible to know
whether those cases were isolated glitches or part of a more troubling
If Bush’s totals weren’t artificially enhanced, they would represent
one of the most remarkable electoral achievements in U.S. history.
In the two presidential elections since Sen. Bob Dole lost to Bill
Clinton in 1996, Bush would have increased Republican voter turnout
nationwide by a whopping 52 percent from just under 40 million votes for
Dole to just under 60 million votes for the GOP ticket in 2004.
Such an increase in voter turnout over two
consecutive election cycles is not unprecedented, but has historically
flowed from landslide victories that see shifting voting patterns, with
millions of crossover voters straying from one party to the other.
For example, in 1972, Richard Nixon increased
Republican turnout by 73.5 percent over Barry Goldwater’s performance
two elections earlier. But this turnout was amplified by the fact that
Goldwater lost in 1964 to Lyndon Johnson by about 23 percentage points
and Nixon trounced George McGovern by 23 percentage points.
What’s remarkable about Bush’s increase over the
last two elections is that Democrats have done an impressive job
boosting their own voter turnout from 1996 to 2004. Over this period,
candidates Al Gore and John Kerry increased Democratic turnout by about
18 percent, from roughly 47.5 million votes in 1996 to nearly 56 million
What this suggests is that Bush is not so much
winning his new votes from Democrats crossing over, but rather by going
deeper than many observers thought possible into new pockets of dormant
But where did these new voters come from, and how
did Bush manage to accelerate his turnout gains at a time when the
Democratic ticket was also substantially increasing its turnout?
While the statistical analysis of these new voters
is only just beginning, Bush’s ability to find nearly 9 million new
voters in an election year when his Democratic opponent also saw gains
of about 5 million new voters is the story of the 2004 election.
Exit polls also suggest that voters identifying
themselves as Republicans voted as a greater proportion of the
electorate than in 2000 and that Bush won a slightly greater percent of
the Republican vote.
The party breakdown in 2000 was 39 percent
Democrats, 35 percent Republicans, and 27 percent independents. In 2000,
Bush won the Republican vote by 91 percent to 8 percent; narrowly won
the independent vote by 47 percent to 45 percent and picked up 11
percent of the Democratic vote compared with Gore’s Democratic turnout
of 86 percent. [See
http://www.cnn.com/ELECTION/2000/epolls/US/P000.html for details.]
According to exit polls this year, the turnout
broke evenly among Democrats and Republicans, with about 37 percent
each. Independents represented about 26 percent of the electorate. Kerry
actually did better among independents, winning that group of voters by
a narrow 49 percent to 48 percent margin.
However, Bush did slightly better among the larger
number of Republican voters, winning 93 percent of their vote, while
matching his 2000 performance by taking about 11 percent of the
While this turnout might strike many observers as
unusual in an election year that witnessed huge voter registration and
mobilization efforts by Democrats and groups aligned with Democrats, the
increased GOP turnout does seem to fit with the campaign strategy
deployed by the Bush team to run to the base.
From the start of the 2004 campaign, political
strategist Karl Rove and the Bush team made its goals clear – maximize
Bush’s support among social and economic conservatives – including
Evangelicals and Club for Growth/anti-government conservatives – and
turn them out by driving up Kerry’s negatives with harsh attacks
questioning Kerry’s leadership credentials.
This strategy emerged from Rove’s estimate after
the 2000 election that 4 million Evangelical voters stayed home that
year. The Bush/Rove strategy in 2004 rested primarily on turning out
that base of support.
But, even if one were to estimate that 100 percent
of these Evangelical voters turned out for Bush in 2004 and that 100
percent of Bush’s 2000 supporters turned out again for him, this still
leaves about 5 million new Bush voters unaccounted for.
Altogether, Bush’s new 9 million votes came mainly
from the largest states in the country. But nowhere was Bush’s
performance more incredible than in Florida, where Bush found roughly 1
million new voters, about 11 percent all new Bush voters nationwide and
more than twice the number of new voters than in any other state other
Bush increased his turnout in all 67 Florida
counties, marking the second consecutive election in which Bush
increased Republican vote totals in all Florida counties, and overall
achieved a 34 percent increase in Florida votes over his 2000 total.
Since Bob Dole’s 1996 turnout of 2.24 million
Florida votes, Bush has increased the GOP’s performance in the state by
an astonishing 74 percent. Making Bush’s gains even more impressive,
Kerry also saw gains in all but five Florida counties and in 22 counties
earned at least 10,000 more votes than Gore earned in 2000.
But Bush’s vote gains exceeded Kerry’s in all the
large counties in the state except in heavily Democratic Miami-Dade,
where Kerry increased his turnout by 56,000 new votes compared with
Bush’s 40,000 new votes. This Democratic improvement in Miami-Dade seems
to have come in large part from Democratic success in registering new
voters in the county by almost a 2-to-1 margin over Republicans.
In spite of this new-voter registration advantage,
Kerry only earned a 7-to-5 increase of new voter turnout over Bush in
Miami-Dade, a statistical oddity given the fact that Kerry did a better
job than Gore in turning out his Democratic base, earning a vote total
equaling 85 percent of all registered Democrats in the county compared
with Gore’s total in 2000 equaling 83 percent of all registered
In other Democratic strongholds of Broward and Palm
Beach counties, Kerry gained 114,000 new voters, earning nearly 770,000
votes, and bested Bush by more than 320,000 votes. But, this was
actually a modest improvement for Bush over 2000, thanks to Bush’s
increase of 119,000 new voters in these counties, from 330,000 votes in
2000 to 449,000 votes in 2004.
Bush’s performance in these two counties is worth
studying in greater detail. In both counties, Democrats saw a
significant increase in new voter registration since 2000, more than
77,000 newly registered Democrats in Broward and 34,000 newly registered
Democrats in Palm Beach.
Republicans on the other hand only registered
17,000 new voters in Broward and a bit more than 2,000 new voters in
Palm Beach. While both counties saw substantial numbers of new
unaffiliated or third party registered voters, the Democratic advantage
in both counties combined of more than 111,000 newly registered Dems
against fewer than 20,000 newly registered GOP voters, as well as the
voter intensity that these new registration rates usually represent,
suggested that Kerry should have done better than Bush relative to the
Instead, Bush actually increased his vote total in
the two counties by earning about 5,000 more new voters than Kerry.
Beyond southern Florida, Bush took turnout
throughout the state to a new level, testing the bounds of statistical
probability by winning votes seemingly from every corner of the state,
from the panhandle to the Gulf Coast, from the I-4 corridor to the
Atlantic Coast from Jacksonville to Miami.
Another county worth examining in some detail is
Orange County, a swing county home to Orlando in the center of the
state. As in Miami-Dade, Palm Beach, and Broward counties, Democrats
successfully registered substantially more new voters than Republicans,
about 49,000 new Democrats against about 25,000 new Republicans.
These gains broke what was once a statistical tie
in registered voters between the parties, giving Democrats a 214,000 to
187,000 advantage across the county. But Kerry only managed a narrow
countywide victory with 192,030 votes against 191,389 votes for Bush. In
2000, Gore carried the county with 140,115 votes against 134,476 votes
While it's conceivable Bush might have achieved
these and other gains through his hardball campaign strategies and
strong get-out-the-vote effort, many Americans, looking at these and
other statistically incredible Bush vote counts, are likely to continue
to suspect that the Republicans put a thumb on the electoral scales,
somehow exaggerating Bush's tallies through manipulation of computer
Only an open-minded investigation with public
scrutiny would have much hope of quelling these rising suspicions.