The Final National Exit Poll (NEP) is the
smoking gun of 2004 election fraud.
The 2004 NEP had Kerry winning by a steady 51-48% from the 4pm (8649 respondents) to 12:22am (13047 respondents) timeline. But Bush won the 2pm Final NEP (13660 respondents) by 51-48%. In the Final, impossible weights and implausible vote shares were required to force a match to the recorded vote count.
The Final NEP was forced to match the recorded vote by a combination of a) adjusting the "How Voted in 2000" weights to Bush 43/Gore 37% and b) increasing Bush vote shares of returning Gore, Bush, Nader and new voters to implausible levels. But the Bush 43% weighting is impossible since it implies that 52.6mm of the 122.3mm who voted in 2004 previously voted for Bush in 2000. This is 2.1mm more than his 50.5mm recorded vote. And it’s 3.9mm more than the 48.7mm Bush 2000 voters who were still living in 2004.
Since impossible weights were required in order to match the recorded vote, the only logical conclusion is that the recorded vote must also have been impossible.
The NEP has always been matched to the recorded vote on the assumption that it is a perfectly accurate count. But in every election millions of votes are cast which are never counted and most of them (70-80%) are Democratic. Therefore, the recorded vote never reflects the true vote. Uncounted votes are a combination of spoiled, absentee and provisional ballots. Florida 2000 and Ohio 2004 are obvious examples.
If we assume a 95% turnout (46.27mm) of Bush 2000 voters in 2004, the Bush weighting is reduced to a feasible 37.8% (46.27/122.3).
The True Vote Model (TVM) used 38.2 Kerry/37.8% Bush weights applied to the 12:22am NEP timeline vote shares and the Greg Palast/Census estimate that 3.45mm votes were uncounted. The TVM determined that Kerry won the popular vote by 66.1-58.4 million (52.6-46.4%) and the electoral vote by 336-202.