Reply #96: Did Bush Commit Fraud? Does a bear go in the woods?

[glengarry]

He sure did. And you warned about it before the 2006 midterms:
http://www.democraticunderground.com/discuss/duboard.ph...

I know you like numbers, so take a look at this:
2004/2006 Election Fraud Analytics: Response to the TruthIsAll FAQ
Last Updated April 23, 2007 by TruthIsAll

http://www.geocities.com/electionmodel/TruthIsAllFAQRes...

Part I contains a comprehensive statistical analysis of 2004/2006 pre-election and exit polls. The 2004 National Exit Poll had Kerry winning by a steady 51-48% from 4pm (8649 respondents) to 12:22am (13047 respondents). 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 True Vote Model determined that Kerry won the popular vote by 66.1-58.4 million (52.6-46.4%) and the electoral vote by 336-202.

In the 2006 midterms, the Democrats gained 31 congressional seats – a solid victory. But they actually did much better than that. A regression trend analysis of 120 pre-election Generic polls (all won by the Democrats) projected they would win by 56-42% – and gain more than 40 seats. The National Exit Poll at 7pm (55D-43R) confirmed the pre-election trend: a Democratic tsunami was taking place. But at 1pm the next day, the Final NEP used implausible weights and vote shares in order to once again force a match to the recorded vote count. The Democratic margin was cut in half to 52-46%. The fraud resulted in the loss of 10-20 additional seats.

To test the robustness of the models, a comprehensive set of sensitivity analysis tables display the effects of incremental changes in input assumptions on the true popular/electoral vote (i.e. pre-election polls: undecided voter allocation; post-election exit polls: timelines, demographic weights/vote shares; voter turnout; cluster effect, uncounted vote shares, etc.) Links to several Excel-based models are included for readers interested in using their own scenario input assumptions and then ask “what-if” to estimate the true vote.