Re: Small Vote Manipulations Can Swing Elections

From: Jerry Lobdill <lobdillj_at_charter_dot_net>
Date: Fri Sep 15 2006 - 18:16:07 CDT

At 02:00 PM 9/15/2006, you wrote:

>Message: 1
>Date: Wed, 13 Sep 2006 14:32:16 -0500
>From: Jerry Lobdill <<>>
>Subject: [OVC-discuss] New Paper on Audit Design
>To: Open Voting Consortium List
>Message-ID: <
>Content-Type: text/plain; charset="us-ascii"; format=flowed
>I have published a new paper on election audit design. You may download it at
>People may also want to read this paper published at
>Small Vote Manipulations Can Swing Elections


These students show that they are thinking about the issues and that
they probably had a late night bull session about it and decided to
get it published. They set up an improbable constraint on the
attacker (she can only change a fixed number of votes at all DREs in
the election.) and they give her an implausible motive (she wants to
alter the election outcome without being detected, but attacks all
DREs in the election.). They choose the 2000 presidential election
for their story. Given the margins in Table 1 of their paper, simple
Excel spreadsheet math shows that this particular election could have
been reversed in the manner they describe. Ho Hum. The margins are so
incredibly small that their story is supported by the data.

But... why would an attacker do that? And why would she have to
change a certain number of votes in every DRE?

The story was written in October 2004 and did not have the benefit of
the Brennan threat analysis. It defies credulity that an attacker
with access to all the DREs in FL, NM, WI, IA, and OR would create
such an attack. Knowing, as she did, that the election was going to
be a squeaker, yet not knowing how close it would be she would not
plan to use her power to distract or attack all machines with small
vote changes. She'd plan to have GWB win decisively, because to err
by changing too few votes would cause GWB to lose. She'd attack as
few machines as possible rather than attack them all.

Jerry Lobdill

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 Sat Sep 30 23:17:04 2006

This archive was generated by hypermail 2.1.8 : Sat Sep 30 2006 - 23:17:08 CDT