Error Rates in the 2004 Election Results

News Flash:

Not only are the politicians screwed-up-cheaters, but the voting methods we use make determining a clear winner in close races impossible.

My good friend Dr. Philip Howard of the University of Washington and his colleagues at www.campaignaudit.org have produced a study that examines the margins of victories in the Presidental, Senatorial, and Gubenatorial races in the 2004 election, and compares them to the known "error rates" of various voting technologies. The surprising conclusion is that, since all voting technologies "lose" a certain number of votes (some more than others, between 1%-2%) , several close races can never be decided with certainty, since the margin of victory was within the error rate. The authors of the study conclude that:

There are, however several states where the margin of technology error was
higher than the margin of victory for a presidential candidate (Iowa, New Mexico, New Hampshire), the margin of victory for a gubernatorial candidate (Washington), or the margin of victory for a Senate candidate (Florida, Kentucky, South Dakota).

This is a non-partisan study, but only one of the "questionible" races put a Democrat in office (Kerry, narrowly, in New Hampshire). No comment on that, except to say that Mr. Bush seems to pull out all of the close ones . . . Hmm.

It boggles the mind, that in 2004, we can't count votes accurately, and we don't seem to care. The question that I'm left with after reading this study is: Fraud aside, what can we do to make vote counts more accurate? Is there a magical system that everyone should use? Should we, at the very least, mandate the systems with the smallest error rates: optical scan (1.2%) and Datavote punch cards (1.0%) ? More hanging chads anyone?

Tin Foil Out