TAR--audits. urgently need help to get law.

From: Charlie Strauss <cems_at_earthlink_dot_net>
Date: Thu Nov 16 2006 - 21:46:24 CST

Kurt, (or anyone). I urgently need your help turing the Targeted
Audit Recount into law.

In New mexico were writing a law to introduce more rigorous recount
procedures. The current bill is enclosed.
But we just got the invitiation to include a Targeted Audit in to the
bill by the subcommittee who will be recommending the bill but we
have to get the amendments in by tomorrow (friday).

First is there any attempts to write TAR as legislation alraedy

Second TAR has one very serious problem that is keeping me from
finishing this amendment.

Unlike a randomly sampled recount, you can't rigorously use the
results of a TAR to estimate how many machines are making errors. If
you can estimate how many machines in the full population are making
errors then you can construct a resonable argument about how likely
it is that there could be enough errors in the election to change
it's outcome based on the sample. From that you can either
iteratively augment the sample size or decide a full recount is needed.

With TAR you are stuck. You can see that an error was made but you
can't easily estimate how big the total error is and therefore you
can't decide what to do next.

You could throw up your hands and say well if there is even one
ballot different that you can't explain in the recount that all the
machines must be checked. You could say that, but you would never
get your law passed, so there's no point. Gotta keep it really simple.

So what is the best way to use non-random sampling for decisions?

Here's the draft of what stands a high chance of becoming law in NEw

The basic law simply says this:
enough machines will be sampled (hand counted) such that there is a
90% chance of detecting whether an election outcome would be reversed
by bad machines. ("machine" here means precinct-machine-
equivalents: batch counted absentee ballots are divided into sub-
batches of the size of a typical precinct ballot count). If a
mismatch between the hand count and the electronic count is found
then an election commission will decide how to expand the recount:
their criteria is to keep expanding until they estimate their is less
than a 10% chance the election would be reversed.

The groovy thing about stating the problem this way is that the
number of machines that are tested is not fixed but varies with the
margin of the election seen in the canvas. (more specifically, the
closest federal, state, and local election is consider as the
operative margin when deciding how many machines to recount in
federal, state, and local races). So if the margin is 2% in the
tightest race then in NM we would recount something like 33 machine
to achieve a 90% confidence of detecting a bad machine if there were
enough bad machines to change the election outcome.

Your comments on this are welcome. The one thing we are not happy
with is the delegation of decision making authority to an election
commission. The problem this is trying to deal with is that there
are so many kinds of errors and scenarios to deal with (e.g. only
expanding the recount in one county is warranted) that codifying them
all seemed worse that delegating to a commission. So if you have
suggestions to make there, don't argue over the statistics as were
not going to change that now, but rather tell us how to codify the
expansion process to avoid the commission.

But really the urgent question is how to add a TAR to this and know
what to do about it's results.

On Nov 7, 2006, at 10:36 AM, dr-jekyll@att.net wrote:

> Here's what I have in my Targeted Audit Recount (TAR) for the
> candidate for office contests. The ballot question contests have
> similar rules:
> Candidate For Office Contests:
> Number of Precincts to be audited based on Contest Size:
> Large – State-wide
> contests 10
> precincts
> Medium – Congressional or State Senate
> size 5 precincts per contest
> Small – State Representative, County-Wide, and
> Municipal 5 Precincts/County
> Precinct Allocations:
> 20% – Winning candidate
> 20% – Allocated to all the losing candidate(s) with less than 10%
> of the vote count.
> Chosen via pro-rata, consensus, or random selection process, such
> as drawing lots in the presence of the candidates, of those
> precincts nominated by members of this group.
> Remaining precincts to losing candidates with 10% or more of the vote.
> Chosen via pro-rata, consensus, or random selection process of
> those precincts nominated by members of this group. In most
> elections, this will be one candidate.
> If evidence of fraud is found, that particular allocation gets an
> addition allocation 5 times greater. I have the initial
> allocations small in order that the people doing the recounts are
> attentive. I believe exhaustive random audits are too tedious.
> That kind of boredom dulls the people's senses.
> Also, it has time limits in order to be sure it is done within a
> few weeks after the election.
> --
> Kurt
> This email sent using 100%
> recycled electrons.
> -------------- Original message from SomeThoughts@aol.com:
> --------------
> lobdillj@charter.net writes:
>> My proposal included the provision that each losing candidate
>> would be allowed to select one precinct from each county in her
>> election district for the audit. But the audit would otherwise be
>> by random selection.
>> The first to propose this was Saltman so far as I know.
> I would propose that any losing candidate within 10% of the winner be
> allowed to pick 1% of precincts/absentee grouping.
> This would be on top of a random audit.
> Jim
> From: SomeThoughts@aol.com
> Date: November 7, 2006 8:57:33 AM MST
> To: ovc-discuss@listman.sonic.net
> Subject: Re: [OVC-discuss] Fwd: Question re. audits in Wisconsin
> _______________________________________________
> OVC-discuss mailing list
> OVC-discuss@listman.sonic.net
> http://lists.sonic.net/mailman/listinfo/ovc-discuss
> _______________________________________________
> OVC-discuss mailing list
> OVC-discuss@listman.sonic.net
> http://lists.sonic.net/mailman/listinfo/ovc-discuss

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 Thu Nov 30 23:17:07 2006

This archive was generated by hypermail 2.1.8 : Thu Nov 30 2006 - 23:17:19 CST