(A Sobering Tale by Lindeman) H.R. 811: Fact & Friction -- Part II

[Wilms]

An analysis of the of the limitation of the methodology used by the Brennan Center in support of Holt's H.R. 811's tiered audit protocol. Lindeman sharpens his pencil. The Brennan's should, too.

Read it. Learn.

(Exit Poll True Believers, there will be a test on this.)


"You might wonder why it could possibly be a good idea to allow inadequate audits in some races while mandating needlessly large audits in others. (Even if you think there is no such thing as a “too large” audit, the misallocation of resources ought to trouble you.)" - Lindeman

H.R. 811: Fact & Friction -- Part II

Guest blogged by Mark Lindeman, Ph.D.*

March 10, 2007

In Part I of this series, Howard Stanislevic pointed out that the audits mandated by H.R.811 (as written) would be far too small to confirm the outcomes of some close elections – while being far larger than necessary to confirm the outcomes in other races.

snip

I like Rush Holt, and I don’t fault him and his people for trying to figure out how to pass a bill. And I think the 3% to 10% audits are not only “better than nothing,” but much, much better than nothing. Still, as compromises go, this one is strange. Not only do some people feel it is too soft, and some people feel it is too tough – but the numbers say that it is too soft and too tough. That’s interesting, but not really in a good way. I think we can do better.

snip

The trouble is, even by the Norden letter’s calculations, H.R.811 clearly does not deliver high confidence in the outcome of every race.

snip

Even if we set aside very close races, the confidence problem is actually worse than indicated by the Norden letter. Unfortunately, the letter’s reference to “imagined typical” districts is only too accurate. In real life, congressional districts don’t have “precincts of roughly equal size.” That wouldn’t matter if we could expect that any vote miscount would be randomly scattered across precincts regardless of their size. But what if an attacker were able to target the largest precincts? Then fewer precincts would have to be miscounted in order to reverse the election outcome, and it would be even harder for a random audit to detect the fraud.

snip

It's one thing to complain that someone else's proposal isn't good enough, but the real question is: can the country feasibly do better? Howard and I are convinced that it can. Since H.R.811 audits too few ballots in some races and more than enough in others, it should be possible to attain more confidence in election results at little or no additional cost. To test our reasoning, we decided to examine all federal elections in the last three cycles (2002 through 2006) -- the presidential race, elections for all 100 Senate seats, and almost 1300 House races. We looked at the consequences of H.R.811's quirky allocation of audit resources, and we explored some alternatives. We will present the results of this analysis in Part III of this series.

snip

http://e-voter.blogspot.com/2007/03/hr811-fact-friction-part-ii_10.html


ER Discussion on Part I:

(A Sad Tale by Stanislevic) H.R. 811: Fact & Friction -- Part I
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x468830

[BeFree]

1. All too complicated

Lets make this easy, real easy.

Take the paper ballots and make two piles (or three depending on how many candidates are running) place votes for candidate A in one pile and B in the other.

Look at the size of the stacks and compare to the machine count. If around 50-50 hand count all the ballots.

Too simple? Yeah, for some.

[Bill Bored]

2. But what happens when...

1. You have more than one race on the ballot?
2. You have more than one precinct?
3. You have more than one county?
4. You have more than one state?
5. You have more than one district?
6. You have districts that cover parts of more than one county?
7. You have counties that cover parts of more than one district?

Oh, never mind. I was thinking of an ELECTION. Sorry for the digression. What were you saying about those ballots again?

[BeFree]

6. Easy

1. Each race is counted the same way
2. Each precinct is counted the same way

And so on.

I guess you are not able to enjoy the simplicity and you think that machines are to be trusted and accepted to count the votes? That humans have no place in counting the votes? Well, sport, that's the mindset that got us into this mess to start with.

[Bill Bored]

9. I guess you haven't read HR811 or the article cited in the OP. nt

[Bill Bored]

10. So let's say you have 50 races on the ballot...

...at every precinct, you sort them into piles 50 times so you can count each race by measuring the height of the 100 piles? Is that it?

You shuffle the ballots 50 times on election night to make piles.

And how do you count the vote-for-more-than-1 races? Do you make a pile for every possible combination?

Come on, just try to think out of your little box for a little while.

No one does HCPB the way you guys are suggesting. If they do, it's because they only have ONE race on the ballot to count.

You don't even have a step where you aggregate the precinct totals! So all you will end up with is a list of precincts with the winners and losers of each one. Like some mini Electoral College voting for Dog Catcher!

If you don't have anything to contribute to a discussion regarding HR 811 or the article in the OP, I'm asking you nicely to find another thread.

This dumb bill may pass as written in part because there is not enough intelligent public DISCUSSION going on about how to improve it. If we can't do that on DU anymore, the way we USED too, then it's over for this place.

[BeFree]

15. Too easy for you?

Yes, you make fifty stacks if that's what you have to do. Or a hundred, whatever.

We are talking about audits, right? Audits of the machine scanned ballots?

So the machine says 60-40 and if the stacks are close to 60-40 the machine can be determined to be, at least, close. If the machine says close to 50-50 you hand count all the ballots for that race. Easy to do if they are already stacked... in that case you just count the number of ballots in each stack.

For some the simple idea of HCPB seems impossible. That's just plain wimpy. The whole thing is far too important to leave in the hands of a few programmers. But that's just what 811 is reaching for and no amount of finagling is ever gonna change that.

As for the way things USED to be, all I can say is look in the mirror.

[Febble]

7. self-delete

(posted to ignoring poster, LOL)

[kster]

3. Do they ever quit, ITS OVER, we are ON TO THEM

THE F*CK*NG game is OVER, We will be Policing the Ballot Count from NOW ON, THE F*CKING ELECTRONIC SECRET VOTE COUNTING GAME, is OVER.

They can word this b*llsh*t any way they want, but it is over, what the f*ck don't they get about this, IT IS OVER!!!

WE KNOW WHAT THEY (THE POLITITCIANS) ARE DOING, ITS OVER ASSHOLES!!! TRY US!!!

:)

[Wilms]

4. Great news! Do you have a link? n/t

[kster]

11. Here ya go

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x469036#469042 :)

[Bill Bored]

12. I guess this is what's meant by "circular reasoning."

If that's all you're willing to contribute to this discussion, then say Hello to your NEW voting system:

:banghead:<http://thomas.loc.gov/cgi-bin/bdquery/z?d110:h.r.00811:>

[kster]

13. Heres a breath of fresh air for ya

no double talk, just plain old fashoin COMMON SENSE, we need to get this to Clinton and Holt!!

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x469062

[Bill Bored]

14. That's better. So, how do we get it passed?

Will Shrub's successor veto it?

Have you read it? Is there anything silly in it like in HR 811 that no one has noticed yet, and won't until it's too late?

Details, kster! We need DETAILS!

[Wilms]

16. Gee. One of the people who fought for that bill...

...Jerry Lobdill...
http://voterescue.org/hb3894.html

...devised an auditing scheme.

.pdf
http://vote.nist.gov/Considering-Vote-Count-Distribution-in-Designing-Election-Audits-Rev-2-11-26-06.pdf

How do you explain that?

Is it possible that one can walk and chew gum in the same time frame?


Also, while this TX bill is great news, in the past you've recommended against reliance on politicians because they were "elected" using the e-voting we reject. Meanwhile, on this thread, you said, "we need to get this to Clinton and Holt!!", refering to the TX bill.

Why would you bother?

[Bill Bored]

20. No conflict there at all. A 100% audit is very effective.

Much more so than the HR811 audit. Not that much more than a proper audit. So I think if by some chance this doesn't pass, the best alternative is a proper audit.

[Wilms]

21. I was asking kster to opine on the subject.

I had a feeling you would grok it.

But in asking kster for a perspective on it, I was hoping to foster, well, perspective.

Should I have used ALL CAPS!? :shrug:

[kster]

23. I'm willing to fight so that you, BillBored, can view ALL the votes being counted

why would you work against me? You should be with me, Godamn it.

[Febble]

24. He is with you

and he's NOT working against you.

The enemies of fair elections are not posters on this forum.

[kster]

25. Is Bill from you Country, or is he from America? nt

[Febble]

26. Well, he sounds like an American to me

a New Yorker, I'd say.

[Bill Bored]

32. Work? What kind of work?

Are you in TX supporting this bill? Did you attend a hearing there? Is there such a bill in your state?

And I still don't understand how the glass ballot box works when you have more then one race on the ballot. If you can explain it, I might understand it. Also, you have to explain how the precinct totals will be added up.

You can't just take a precinct with a bigger bunch of ballots for Candidate A than for Candidate B and say that the Candidate A won the precinct. You have to add up every vote in every precinct!

I appreciate the transparency metaphor, but to get it to work, you need to do more than just put ballots in a see-through box. You have to count all the votes in every race on the ballot and add up all the precincts.

I'm on your side, but if this bill in TX doesn't pass, I'd like to know what your alternative would be?

[kster]

33. Do you know why there are 3,168 US Soldiers Killed, 23,677 Seriously Wounded

Because we did not Hand Count the Ballots, thats 3168 Americans Killed as March 1st, 2007, Why? because in 2000 we (Americans) were asleep at the wheel, Now we (Especially ER people) should KNOW BETTER, for some reason there are people in here that want to let, these machines continue to count our ballots. Why?

Because its TO HARD TO HAND COUNT, OR IT IS IMPOSSIBLE TO COUNT A BALLOT WITH 20 RACES ON IT, You GOT TO BE KIDDING ME, You got 3,168 US Soldiers Killed, 23,677 Seriously Wounded, and it's TO HARD to Hand Count Paper Ballots, PLEEEEASE.

Finding a way to Count the Ballots by Hand, is the least we can do for the families of the Killed and Wounded in this war that happened, because, the ballots WERE NOT COUNTED BY HAND, in the first place.

http://usliberals.about.com/od/homelandsecurit1/a/IraqNumbers.htm

Statement from 2000:

Don’t set aside votes just because it’s hard to count

I agree with something Governor Bush said last night. We need to come together as a country to make progress. But how can we best achieve that? Our country will be stronger, not weaker, if our next president assumes office following a process that most Americans believe is fair. In all our hands now rest the future of America’s faith in our self-government. The American people have shown dignity, restraint and respect as the process has moved forward.

This is America. When votes are cast, we count them.

(((((We don’t arbitrarily set them aside because it’s too difficult to count them)))).

Two hundred years from now, when future Americans study this presidential election, let them learn that Americans did everything they could to ensure that all citizens who voted had their votes counted. Let them learn that democracy was ultimately placed ahead of partisan politics in resolving a contested election. Let them learn that we were indeed a country of laws.

Source: Speech on primetime national television Nov 27, 2000


http://www.issues2002.org/2008/Al_Gore_Principles_+_Values.htm#Florida_Recount

[kster]

22. Oh my, you said

Also, while this TX bill is great news,

(((in the past you've recommended against reliance on politicians because they were "elected" using the e-voting we reject))).

Meanwhile, on this thread, you said,

"we need to get this to Clinton and Holt!!"

refering to the TX bill.


----------------------------------------------------------------
This is what is known as SARCASM, Clinton and Holt are part of the establishment, and RANDOM AUDIT is as big of a part of the secret vote counting SCAM, as the secret vote Counting Machine themselves, you know that as well as I do, Its all about preventing the people from VIEWING the VOTES BEING COUNTED (PERIOD)
------------------------------------------------------------------


Do you grok? :) I know for a fact that you do.

[Febble]

5. Excellent

Your country needs wonks.

[Febble]

8. Come on guys!

This one needs recs!

[Awsi Dooger]

31. Like Maxwell Smart, I missed it by that much

Tried to recommend but got an error message that it was limited to within 24 hours of the thread being posted. Well, I learned that anyway.

This thread probably got ripped off, posted a few hours before Daylight Savings Time.

[Wilms]

17. Report Finds Disparity in State Approaches to Post-Election Audits

electionline Weekly – March 8, 2007
electionline.org

I. In Focus This Week

Report Finds Disparity in State Approaches to Post-Election Audits

More states seek to build voter confidence through audits, but methods vary

By electionline.org staff

A comprehensive new case study of post-election audits by electionline.org finds that while manual audit use is on the rise nationwide, the methods, sample size and remedies for disparities are markedly different across state lines.

snip

But how states conduct those audits varies significantly. For example, while California auditors analyze ballots from 1 percent of precincts, Connecticut counters examined votes from 20 percent of all precincts in the state. And while Wisconsin makes voting-system vendors accountable for problems revealed in audits, a number of states expand audits and leave broad discretion to chief election officials to determine how to proceed in case discrepancies are found.

snip

The case study, the 17th in a series of briefings on election administration issues, focuses on five states that use different approaches for conducting audits.

snip

The report provides a snapshot of audit rules and procedures in 18 other states, including states that perform other types of post-election reviews that do not involve manual counts as well as a review of what changes federal legislation could have on auditing processes nationwide.

It is available online here.

pdf
http://www.electionline.org/Portals/1/Publications/EB17.pdf

snip

http://www.electionline.org/Newsletters/tabid/87/ctl/Detail/mid/643/xmid/241/xmfid/3/Default.aspx

[OnTheOtherHand]

18. the good, the bad, and the ugly

You almost have to laugh about Arizona, where it seems that essentially one party can block an audit by just not showing up. And then there are states like Minnesota where so many people work so hard to make the system work.

[Wilms]

19. My favorite is NV where they found the process “tedious and error-prone”.

Democracy is hard work. Real hard work.

But then, as a poster here, you knew that. :D

[Land Shark]

27. Why publish, in the United States Code, the method of audit investigation of the bad guys?

It's not just a matter of tweaking audit formulas. There are huge problems with the entire approach of "auditing" in the first place, like announcing how accuracy or fraud will be tested. That helps a potential wrongdoer a great deal in terms of knowing what he or she can get away with. Do we do this with any other types of investigations or audits? Does the Dept of Education go in and audit schools and tell them in advance everything they will be tested or evaluated on? Do we instruct the FBI by statute how to investigate bank robbery?

Has there been an adequate answer to Roy Saltman's suggestion, about 30 years old now, that a level of confidence be specified and that the auditors design the specifics of that based on the unique attributes of the district they are auditing?

[OnTheOtherHand]

28. it's hard to know what to make of this post

On the first point: Is there some particular specification that you think should be avoided? For instance, is it a bad idea to specify that audits must be random, because that gives the bad guys valuable hints?

On the second point, I assume you're aware that Stanislevic has been arguing for confidence-based audits for quite some time now.

[Wilms]

29. I know for a fact that Stanislevic is arguing based on a fixed confidence level.

Like 99%.

And he did a lot of digging into Saltman's work. There are a few articles about auditing on the right hand column of the blog.

Put a comment on the blog. He'd probably answer.

[Bill Bored]

30. Answer to Saltman's suggestion has been,

"Ignore it and it will go away."

Fortunately, there are some in the "community" who are pushing for it against all odds, or should I say against all probabilities?;)

If it's ever seriously considered, it won't be because of anything noble like ensuring election integrity, but more likely because it can be cheaper than doing it wrong as Holt is now proposing. I guess that makes it a "win-win", huh?;)