Re: A 3-Step Audit Protocol w/ 99% confidence

From: Kathy Dopp <kathy_dot_dopp_at_gmail_dot_com>
Date: Fri Jan 26 2007 - 19:43:16 CST

> Date: Fri, 26 Jan 2007 17:17:09 -0600
> From: Jerry Lobdill <>
> Subject: Re: [OVC-discuss] [EILeg] [ei] Re: A 3-Step Audit Protocol w/
> 99% confidence

> characterization, because Howard and I both came up with this method
> independently, and I am very familiar with it.

Can we please stick to the facts Jerry? You and Howard are again
claiming credit for my work that I publicly released in July 2006
before either you or Howard ever wrote anything publicly on the
subject and which you obviously read before you announced your skill
in this area, since you were in my work group for a while until we
ejected you after you wrongly claimed credit for Frank Stenger's work.
You and Howard are very skilled at replicating others' work shortly
after it is publicly released without crediting the originators.

However, let's give credit where credit is due (as I always do), since
it was the Brennan Center and Roy Saltman who first came up with the
idea of maximum wrongful vote shift per vote count, which I found in
their appendices and which I first thought of correctly applying to
calculate the number of corrupt vote counts to detect for election

Perhaps you could explain the concepts better if you knew more about
its origins. Roy Saltman and the Brennan Center both explained the
concept of Maximum Vote Shift Per Vote Count very well which is where
I got the idea and used it to correctly calculate the amount of
corrupt counts to detect for election audits by using it with the
margins between candidates:

Maximum Vote Shift

= The Assumed Maximum Percentage of Votes That Could be Wrongly
Shifted from One Candidate to Another, within any one vote count,
without being immediate suspicious.

It's that simple. It applies to ANY type of vote count, not just
precinct counts.

What this logically means, is that the audit is designed to detect any
miscount that could wrongly alter an election outcome in a way that is
not suspicious;

and that any vote counts that look suspicious due to
having more margin swing than that of your assumed maximum vote shift,
must be automatically included in an election audit sample in addition
to the ones determined from the numerical method that Frank and I give
in our paper, or the estimation formula by Ronald Rivest, that I
explained explicitly how to apply to elections in my recent paper of
Jan. 17th

In other words, any vote counts exhibiting a margin shift of two times
the assumed vote count shift in relation to a calculated measure such

1. previous similar elections or

2. exit or opinion polls, or

3. partisan active voter registration measures

are automatically added into the pile of audited vote counts.

There are no holes or weaknesses in this method of calculating audit
amounts. The assumed maximum vote shift is not remotely analogous to a
"fudge factor" and to claim that it is a "fudge factor" is a gross


OVC-discuss mailing list
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Tue Jan 1 14:12:50 2008

This archive was generated by hypermail 2.1.8 : Tue Jan 01 2008 - 14:12:51 CST