[governance] Afghan election - fraudulent "audit" protocols preserve initial numbers

[Paul Lehto]

Under the thread title ICANN/Affirmation of Commitments" a subject
came up regarding internet governance, and I replied to show why
internet voting, specifically, can not only never be secure, it can
never be timely investigated and corrected for error given short
statutes of limitation, and because only the "conclusions" or vote
totals are ever disclosed, there's no scientifically rational basis
for confidence in the results whatsoever, given that no data is
disclosed.  They are purely faith-based elections, specifically in a
context where the government that would run such an election has close
to the ultimate conflict of interest given that the elections
themselves determine the government's own power and composition and
they are tabulating the results in complete nontransparency combined
with claims of corporate trade secrecy that are brought to bear if a
court order is sought for forensic computer evidence.  Internet voting
is nothing to bet democracy or governance or even ten Euros on.

In the past I've done a congressional election contest together with
other election contests.  I've found that post-election remedies like
recounts and audits work so infrequently that they can not be
considered to give any significant support to a flawed and
nontransparent first count -- the one that creates the headlines and
the 'sore losers' in any case.

Here's an example, a perfect one, of the kind of BS that goes on with
post-election remedies.  I haven't confirmed the mathematics myself,
but the author, a mathematician named Kathy Dopp who works in the area
of election analysis, states the following. If for any reason it would
turn out to be a mistake, which is possible but not likely, then
consider it a hypothetical example of the kinds of things I've seen
happen in the post-election time frame regarding real recounts and
audits -- which work so infrequently, only twice in 30 years has a
statewide result been changed (the latest being Senator Norm Coleman
losing to Al Franken in Minnesota just recently):
----------------Afghanistan report ----------- from Kathy Dopp ---------------

Someone *really* needs to contact someone in the US Department of
State or contact someone who can reach President Obama because the
Electoral Complaints Commission (ECC) and the Independent Election
Commission (IEC) are using fraudulent methods of analyzing the Afghan
election audit results that ensure that the initial reported margins
never change regardless of the amount of margin error found in the
audit sample. They are doing this by adding up all the error found in
all candidates' initial counts and subtracting that total share from
*all* the candidates' vote shares and reporting that (unchanged) vote
share (after you renormalize it to add up to 100%) for all candidates.

In other words the methods being applied by these election
organizations *always* preserve the exact same ratios of candidate
shares relative to each other as the original shares were before the
audit, even when there is more than enough error to change the
election outcome (or reduce Karzai's share below 50% and trigger a
top-two runoff).

Here is a short math proof that the method they are using *never*
changes the original election outcome regardless of the error since
they subtract the exact same percentage from every candidates' vote
share.

In other words, if there are three candidates with vote shares
 a%, b% and c% respectively (where a + b + c = 1)

and I subtract x% (no matter what x is) from every candidate's vote
share, then I get

a(1-x)%,  b(1-x)%  and c(1-x)%

when I total these I get (1-x)(a + b + c)%

and then add up all the candidates' shares and divide by that number
I get the new vote shares that add up to 100% of:

a(1-x)/(1-x)(a+b+c)% which reduces to a%

b(1-x)/(1-x)(a+b+c)% which reduces to b%

c(1-x)/(1-x)(a+b+c)% which reduces to c%

These are the exact same initial candidate shares that we started
with.  So the *new* method is a hoax, meant to fool the Afghan and US
public because if you incorrectly subtract the same amount x% from
every candidate's vote share then the candidates will *always* have
the same proportion of votes relative to each other that they had
before you subtracted x%.

That is clearly *not* an honest way to readjust the vote shares in
response to error found in election audits.

----

The State Dept needs to know about this, investigate, and do something
to insist on an honest election process if this is true.
--Kathy Dopp

-- 
Paul R Lehto, J.D.
P.O. Box #1
Ishpeming, MI  49849
lehto.paul at gmail.com
906-204-4026
____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance