Rossi Should Have Won By 100 Votes

Via Daily Pundit we find this article about the statistical analysis given by the GOP to the courts over the 2004 election for governor in the State of Washington:

Without ballots cast by felons, dead voters and non-citizens, Democratic Gov. Christine Gregoire would have lost the 2004 election to Republican Dino Rossi by about 100 votes, according to a statistical report cited by the GOP in its legal challenge to Gregoire's victory.

All well and good, but will it matter at this point?

But it's uncertain if the courts will buy the statistical arguments, which are a key to the GOP effort to overturn Gregoire's 129-vote win. Democrats say state law doesn't allow such arguments.

Weird. I remember them saying that the court challenge couldn't succeed unless it was shown that the improper voting could have changed the result! In fact, the judge said so:

In pretrial rulings, Judge John Bridges has said it won't be enough for Rossi to show the number of improper votes exceeded Gregoire's margin of victory. According to state law and previous court cases, the GOP will have to demonstrate that Gregoire owes her win to illegal votes, Bridges said.

I'll believe the judge and not the Democrats, thank you.

So, how did we get to this +100 number? After all, earlier allegations claimed over 1000 improper votes, but post-analysis the margin changed by approximately 250. Well, they were smart:

Two political science professors—Jonathan Katz of the California Institute of Technology and Anthony Gill of the University of Washington—broke down the data for the GOP.
Both subtracted improper votes from the candidates in proportion to the overall vote each received. When possible, precinct returns were used to establish the pattern; otherwise, county results set the percentages. For example, if Gregoire received 60 percent of a vote in Precinct A and Rossi 40 percent, and there were 100 improper votes from that precinct, Gregoire's total would be reduced by 60 votes and Rossi's by 40.

Sounds fair to me.

Josh Poulson

Posted Wednesday, Apr 20 2005 10:12 AM

Adjacent entries


« Hanoi Jane Unwilling Expectorant Recipient
President Froman »





To track back to this entry, ping this URL:

There are no trackbacks on this entry.


There are 3 comments on this entry.

Katz & Gill's analysis is flawed, however. They treat the illegitimate votes as if they were a random sampling from a given precinct or county. Since the illegitimate votes were undeniably a fraud (whether they were individual frauds or part of an organized effort is unknown), then it is just as likely that the bulk of the illegitimate votes would favor one particular candidate or the other within a given precinct, not simply a random sample. There is no way of knowing if 40% of a precinct's illegitimate votes went to Rossi, or if 90% of them did, regardless of the overall ratios. As such, it is simple speculation and CAN NOT be considered "proof".

But it's also rather interesting that the GOP would use such data as an argument for overturning election results, considering analysis of Florida's 2000 Presidential Election demonstrated that in a state-wide recount, Gore would have been declared the winner.


Posted Wednesday, Apr 20 2005 10:50 AM

Is there some better way for removing the effect of the illegal votes? At least with the discovered absentee ballots they have an idea as to whether that batch of ballots was skewed in any one candidates favor.

I certainly didn't use the word “proof” in this writeup, however, what the court is looking for is based on “preponderance of evidence.” That particular standard is nowhere near the certainty of “proof.” in the mathematical sense.

As for Florida, all the newspapers down there agreed that no matter what method was used to tabulate ballots (that Gore asked for) Bush won. The only method I heard of that gave Gore a win used different standards in different counties.

Josh Poulson

Posted Wednesday, Apr 20 2005 11:09 AM

If Phil is correct, this is an outright invitation to election fraud since you could never, ever prove how a fraudulent voter voted once their ballot was in the vote stream.

Phantom Menace

Posted Saturday, Apr 23 2005 09:22 PM

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)



Affiliate advertising

Basecamp project management and collaboration

Backpack: Get Organized and Collaborate