What would a Smoking Gun actually look like? Questions/Answers

[Land Shark]

A good question to ask an elections official is "what would a vote swapping smoking gun look like, in electronic voting?"

In my experience, they don't have an answer to that question that shows any kind of depth. Any answers one is likely to get will be based on a paper-ballot paradigm (like: when ballots at precinct X exceed signatures in the pollbook by 20%)

In electronic voting, too many votes relative to signatures MIGHT still raise a red flag but voting officials sometimes have powerful rationalization powers at their disposal. More often, they don't know what an electronic smoking gun (or even EVIDENCE) would look like, and they AREN'T LOOKING for it either, of course.

Any nominations for what they should be looking for, electronically speaking?

[mgr]

1. I think some constraints as to what type of technology.

A good indication is, if the equipment has a confirm your vote element, such as in Mahoning, but without the more numerous machine problems (freezing up)is voters taking longer in the booth than would be expected otherwise. Anything requiring detection at a level higher than precinct would be untenable to my mind.

Mike

[BeFree]

2. I've got one

When the presidential votes on election day follow a mathematical formula which shows that for every 5 votes for Kerry there were 6 votes for b**h.

Also, ignaztmouse posted a thread here in November which showed absentee votes far exceeded the percentages of election day votes: sometimes by as much as 15%. In other words, say the vote by absentees was 55% Kerry and the final outcome showed b**h winning by 10%.

[Bill Bored]

3. With or w/o VVPATs?

With VVPATs:

1. A high enough percentage of unaudited VVPATs sufficient to reverse the outcome of the election.
<http://uscountvotes.org/ucvAnalysis/US/paper-audits/Paper_Audits.pdf>

2. A mismatch between the hand count of the VVPATs and the machine count in one, all, or a subset of the machines in the jurisdiction.

With or without VVPATs:

1. An error or other discrepancy in the ballot definition programming showing that votes would actually not have been counted as cast according to the election laws of the state, equal protection, general principles of fairness, etc.

2. Incorrect totals seen by testing of the machines by actually casting the number of votes expected or actually cast in the actual election, comparing the machine totals to hand counted VVPATs or a manual running total recorded as the votes are cast. Such testing should also include but not be limited to undervote testing.

3. An error in the instructions given to voters.

4. A discrepancy between the precinct level machine counts and the counts shown in the central tabulator for the same precincts.

Without VVPATs:

1. Error(s) or discrepancy in the source code would have to be determined by viewing the source code, and/or a digital signature applied to it, to determine what it does, if it's certified, if it's been modified, if it is in fact the same code that was running on election day, and/or early voting days, etc.

2. Oh yeah, massive unexplained exit poll discrepancies!

3. Karl Rove, George W. Bush or Dick Cheney's presence anywhere in the state within a week of the election or prior to the release of any official results!:)

[Melissa G]

57. Now Bill, you keep going on about those massive unexplained exit poll

discrepancies.. :) remember to stay focused on the DRE's and election management systems..:hi:

[Kip Humphrey]

4. For me, when no vote registered for Kerry after selecting "Vote

Democratic Slate" during the 2004 election.

For the MSM, a Press Conference by Rove where he admits stealing the election.

For the Supreme Court, lifetime memberships in Augusta CC.

For the American People, learning what we know.

[Time for change]

7. Like reported here?

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=371211

The appendix that this thread linked to is no longer functional, but I have it in a word document. It is full of examples like that, and similar ones.

[kster]

11. You hit the nail on the head

"For the American People, learning what we know."

And once the Americans learn about it, these companies better take cover.

[Bill Bored]

5. And don't forget undervotes or as Diebold calls them "blank votes"

esp. at the top of the ticket!

Also, anything that looks TOO perfect, i.e., zero undervotes, zero blank votes or zero blank ballots on EVERY machine.

[Land Shark]

6. One elections officials said:

evidence that a vote was not recorded as cast.

Of course, if you think about it, if one could connect a voter with a particular ballot to do this analysis, that would violate the secrecy of the ballot. Thus, this is the type of information not allowed to exist under our system of secret balloting.

[Time for change]

8. I think Clint Curtis' testimony before Conyers' committee comes close

And when the federal investigator who was working on his case was "suicided", that was icing on the cake.

[garybeck]

10. not specific enough

IMHO Curtis testimony is damning, but it is not about any specific election and it is not directly tied to anything. there is no evidence that his program was actually used.

[Time for change]

12. Perhaps you're right

Depending on what the investigation of the "suicided" investigator showed.

[garybeck]

9. how about white stickers found on the ballots, covering the votes?

smells like smoke to me

another one would be if they cracked open the machines and found malicious software on there.

[Land Shark]

17. this is not an electronic smoking gun, per se

[Time for change]

13. I think the Warren County lockdown and the Hocking County cheat sheets

during the recount come pretty close.

So many of these things weren't even investigated, though. It's hard to have real smoking guns when there is no thorough investigation.

[petgoat]

14. Didn't Bob Fitrakis say something about someone who saw somebody

take a whole stack of ballots into some side room in one of the Ohio
counting houses? I wish I could remember. I tried to google about it a
week later and couldn't find it.

[Land Shark]

18. isn't it great that you can do that electronically without anyone "seeing"

[petgoat]

20. Sorry, I was OT n/t

[Melissa G]

15. How about massive vote theft like in Florida that looks very suspicious

but is theft so large that it will surpass the amount that triggers automatic recounts. This is done in so many ways that you cannot prove enough to show in a short period of time, that this corrected information would be enough to change the outcome of the election. Oh wait, that is what happened in Florida..

[Melissa G]

16. a programmer friend of mine says legit programming errors tend to

be wildly large.. small discrepant patterns are likely to be intentional according to my friend

[FogerRox]

19. Sorry I'am late--

The smoking gun would be the Digital ballot images imbedded in proprietary software.

There are the only representation of the voters intent. And yet HAVA mandates that the DRE tape-- end of voting print out-- to be the officals record--- AAARRGGG.

The best place for the smoking gun to be is hidden, protected by Corporate law.

Since the Offical record is a paper "audit" end of voting print out--
YOU CAN GO AHEAD AND ERASE THE DIGITAL BALLOT IMAGES---- AS per section 301 of HAVA--- GET IT? Federal law mandates this----------- in all 50 states.

Federal law mandates that the "copy" paper print out-- is official

And the primary document is erased--- the digital ballot images----

---insert hyperbole--

YOU all get a big fate F

DO I win the prize?

[Land Shark]

25. this comes very close; the only debate is who are the malfeasants

and who are the idiots in passing HAVA like that?

[Bill Bored]

29. As far as passing HAVA,

there is some scuttlebud and debate about how this happened.

I have heard from a reliable source that it was done more or less in the dead of night and the Dems didn't realize that language similar to "voter verified" had been excluded. If you read section 301 it looks pretty good but there's nothing that says the paper audit record has to be voter-verified -- that's the catch:

(2) Audit capacity.--
(A) In general.--The voting system shall produce a
record with an audit capacity for such system.
(B) Manual audit capacity.--
(i) The voting system shall produce a
permanent paper record with a manual audit
capacity for such system.
(ii) The voting system shall provide the voter
with an opportunity to change the ballot or
correct any error before the permanent paper
record is produced.
(iii) The paper record produced under
subparagraph (A) shall be available as an official
record for any recount conducted with respect to
any election in which the system is used.

This is then interpreted as nothing more than a results report, but it also allows mechanical lever machines to be used as long as someone writes down the results on a piece of paper after the election. Of course, this latter point is not discussed very much, unlike the fact that DREs without VVPATs are permitted! It's all about the S P I N !

[Bill Bored]

30. And those images can be altered too, even if preserved. nt

[autorank]

21. Collaborating with the enemy

[eridani]

22. Significantly different results in a parallel election

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=389521

Otherwise, Bill Bored is right--you have to be able to get at the software.

[petgoat]

23. "[election officials] AREN'T LOOKING for [a smoking gun]"

According to TrueVoteMD the election officials in Maryland covered up
the problems with the Diebold machines there.

http://truevotemd.org/Press_releases/html/2005-03-08_Press_Release.html

California has $30 million worth of Diebold machines it can't use, and
another $15 million worth that won't be usable after February.

http://www.insidebayarea.com/localnews/ci_2958901

The smoking gun is the fact that the electronic machines are unauditable
and hackable. Election Officials must entertain a state of denial about
these facts before they can even buy the damn things, and would probably
continue to deny there was an issue no matter what the evidence was.

[Time for change]

24. What about pre-tabulator and post-tabulator results that don't match --

especially if there's a widespread pattern of that.

It seems to me that that ought to be prima facie (excuse my attempted lawyer language) evidence of fraud. The problem is that I've asked about that repeatedly, it seems that almost everyone agrees, and they also agree that those numbers ought to be readily available, and yet nobody can dig them up.

[FogerRox]

26. PRE or POST--

The hardware tends to give it self away--

The hardware of a DRE is built to cheat --a front emd and back end--- each generating results-- like an accountant with 2 sets of books.

Since the ballots are hidden from audit no one can question them. (The digital ballots)

From there the DRE justs spits out a DRE tape and a smatr catd with the desired results---

In other words the scene of the crime must be hidden-- so
Hide the ballots--

[Time for change]

27. Yes, but my understanding was that pre-tabulator totals had to be recorded

So then, if the central tabulator changed the totals it received from the individual precincts, then that could be checked on at a later time. Here's the way that kiwi_expat explained it on Helderheid's thread.

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=387455&mesg_id=388425

[FogerRox]

28. yes you are correct that pre tab are are recorded

A DRE at the end of an election does 2 things--
1)prints out a tally on paper
2)records a tally on a results cartridge.

The results cartridge is then removed-- plugged into the tabulator.

But to me, its makes very good sense to do the crimne in the DRE. and to----
print out a tally on paper & record a tally on a results cartridge, that are both fraudulent. Both match-- so the tabulator is compiling fruadulent results---- the only original evidence is in the DRE--
Those ballot images that are the only record of the voters true intent.

[Bill Bored]

31. Other evidence may be the ballot definition files

which also exist on the server side and may be archived on CDs or other backup media. See post #3.

I do agree that to avoid detection, the fraud would be _executed_ on the client side (DREs). But to be efficient, it would have to _originate_ and _propagate_ from the servers. Otherwise you're talking about hacking one DRE at a time, just like lever machines! This is why e-voting is so dangerous. Remember they're not just tabulators -- they're Election Management Systems!

[FogerRox]

34. And the tabulators cannot be compared to the original record of the voters

intent---ballot images in the DRE.
Assuming the scene of the crime is the tabulators-- (in Ohio, because IIRC only 4 counties used DRES), Then the opscanned paper ballots need to be archived as well as the Punchcards.

But they were already recounted and the recount said Bush won.

Why hack one DRE at a time? You write the ballot on the Tabulator workstation--- right?
But certainly in the case of Ohio it was tabultors not DREs.

[Bill Bored]

36. Exactly. Tabulators=servers; DREs=clients.

You wrote:

"Why hack one DRE at a time? You write the ballot on the Tabulator workstation--- right?"

Right. You write the bogus ballots and tally rules on the server, download them to a zillion DREs, or enough to steal the election, run them on Election Day, Early Voting Days, or both, and send the bogus results back to the tabulator for aggregation. You don't have to change anything after the election because it's already been stolen.

[Time for change]

32. It may be that it makes sense to do it on the DRE side

But in Ohio at least it appears that the good majority of the fraud happened at the central tabulator level or with massive registration purges.

Some studies have drawn the conclusion that vote switching by DREs was not a major cause of fraud in Ohio. One reason for this would be that not much of the Ohio election was run on DREs. There were only about 6 counties in Ohio that used electronic voting. One was Mahoning, which used touch screens, for which there is good evidence of electronic vote switching. Then there were I think 5 other counties that used electronic machines that were not touch screens, and I'm not sure if they were DREs, but I don't think so.

[Bill Bored]

37. The DRE side comes from the server side.

DREs don't just sit there waiting to be programmed individually. They are programmed from the servers, which are the same machines that aggregate the DRE results after the polls close, which we call central tabulators. If you want to steal votes, you don't just switch them on the tabulator where a precinct canvass would turn up evidence that the totals don't match. You do at the precinct itself. But the precincts are "programmed" for lack of a better term, from the server. It's a matter of configuration vs. execution.

As far as Ohio, that's antiquated punch cards mostly. They could switch votes as easily as counting the cards in the wrong machines. Since the cards aren't labeled, who can prove where they belong? And statewide mandatory ballot rotation, even in general elections, makes this possible. Of course, it's all possible on DREs too. With those beasties, you don't even have to use ballot rotation.

[Time for change]

44. That's very interesting

My more general point was that it appears that in Ohio the good majority of electronic fraud was performed centrally. I say that because there have been at least two studies (ESI and Walter Mebane's study which is documented in the DNC report) which have cast doubt on the possibility that widespread fraud was committed by switching votes at the prcinct level. Their studies are based on statistical analysis of Ohio voting patterns, and they may very well be right on this point (I'm not willing to absolutely concede that point, but my guess is that they're correct.)

Would you agree with that?

[Bill Bored]

50. Depends on whether there was fraud or not historically. nt

[Time for change]

53. Could you expand on that?

I'm not sure that's relevant to this point.

[Bill Bored]

55. See posts 8 and 9 in this thread:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x391445

[Time for change]

56. Good point -- I agree that there was some vote switching at precinct level

There definitely was in Mahoning County, which was the only county in Ohio that used touch screens. But I believe that that was only a very small portion of the total fraud in Ohio, and not capable of changing the results of the election by itself.

Same thing with the white stickers on the op scan ballots.

[Land Shark]

33. Many of you tell me WHERE TO LOOK, but not WHAT to LOOK FOR!

[FogerRox]

35. WHAT to LOOK FOR! WHAT to LOOK FOR! WHAT to LOOK FOR!

WHAT to LOOK FOR!
WHAT to LOOK FOR!
WHAT to LOOK FOR!
WHAT to LOOK FOR!

Is what you are asking for, to possibly be forensic evidence? Something tangible-- something I, or you could hold in your hand?

What if we compare down ticket voting patterns with BUSH & KERRY voting patterns.

I was in FLorida- Broward county, and it seemed vote switching went like this: "I voted Kerry & Castor-- it kept coming up Bush & Martinez". (Betty Castor and Martinez ran in the senate race in FLorida). Broward voted with DREs. SO maybe those 2 races would be differnet from the down ticket races--- AND---
If votes were switched, it may show up as--- in the case of a bell curve-- part of the bell curve would seem to move over-- creating the look of a double exposed photo-- 2 bell curves.

Then compare the 2 bell curve jurisdictions with exit polls and then the single bell curve jurisdictions with the exit polls. Is there a pattern?

[Land Shark]

40. such a "double exposure" bell curve should cause a double take!

I assume you knowingly allude to the actual twin peaked curves evidenced in the Snohomish DRE study at www.votersunite.org/info/SnohomishElectionFraudInvestigation.pdf
around page 10-12 or so

[FogerRox]

42. Never heard of that county -- that lawyer-- the State of Washington

never read anything at Votersunite

.................yeah you caught me

I dont think 99% of elections officials look at anything-- they look at their local codes and statues and cover their asses and thats it.

Very rarely would some one do a report like the last page of the Shamos report showing a, by county, undervote rate-- which clearly is machine type driven

I have found in NJ and in my county election officials dont know too much, and the more I dug into NJ election law and HAVA the more I realized we could use this Section 301 thing to beat them over their heads-- and we did---

[Bill Bored]

38. I thought I did. See post #3. nt

[Land Shark]

39. #3 is not bad. What i REALLY want to know, though, is WHAT

ARE OUR ELECTIONS OFFICIALS LOOKING FOR?

As far as I know, they wouldn't know e-fraud if it bit them in the butt. Or would they?

[Bill Bored]

41. I don't know. I'm not an elections official!

It sounds like you've already asked some though and it would be interesting to hear what others have experienced.

But I think everything in post #3 should be codified in law and there should at least be VVPATs if not VVPBs, auditing, manual recounts as necessary, etc. and we shouldn't have to jump through hoops to get them. And elections should not be privatized, even if there is machinery involved.

I think we've tried to do all this in NY, but I'm still looking for loopholes! If there are none, then it's just a matter of enforcement, and I do need to check on what the penalties are because I'm afraid a slap on the wrist won't do.

What I find disturbing is that in some other states, it's basically a lawless process!

[Land Shark]

43. you can have all the law you can imagine, but it won't matter if the

elections officials don't have both the spirit of democracy and the full intent to uphold and defend it against all enemies FOREIGN AND DOMESTIC as they say.

[Melissa G]

46. kickin for Land Shark!

[Bill Bored]

51. Their power is unprecedented whether they know how to use it or not.

E-voting has given these folks the ability to switch thousands of votes with the click of a mouse. This is the problem.

I'd rather have laws than no laws, but still better would be for the power to switch votes to be more widely distributed in the first place. Solely in the hands of each voter would be ideal. Failing that, enough checks and balances n the system to prevent someone else from acquiring it.

This is why your work is so important; it challenges the very notion of concentration of a Constitutional power that was never meant to be concentrated. Something like that.

[tommcintyre]

45. Kick nt

[philb]

47. There were lots of smoking guns in 2004 election; but no officials

who were willing to investigate them.
Is it that too many are spoiled, afraid, corrupt, distracted, or what?

[Land Shark]

48. partial answer: some don't have the slightest idea what to look for

so everything seems to be going fine, to them.... i express no opinion for now whether this portion is 1% or 98% of elections officials.

[Amaryllis]

49. In states where there is a record of party registrations of voters who

voted, look at the patterns of how many voted for the different candidates and see how it compares to party registrations.

[FogerRox]

52. Vaguely similar to what I waas trying to write-- you might have said your

part better though-- yes I Like it--Am

[starmaker]

54. Expectation of fraud in the system

Chuck Herrin talked of this in his presentation to state of nc
It's why system audits are done
Why we have no audits of system components and performance
in a system subject to fraud?
the secrecy of the digitized vote count, unaudited by standard
business practices?

[MadeinOhio]

58. Landshark, this is an excellent question..

I think for Ohio the smoking gun definitely is in the discrepancy between straight-ticket votes and increased votes for Bush.

I would say matter of factly, that according to math anyway, anywhere the votes exceed the number of voters or virtually triple the votes for that demographic fraud is guaranteed.

For instance, for every 3 votes for Kerry there was 6 votes for Bush in some unusual democratic areas of Ohio. It is very hard to establish one, set, consistent pattern.

Rather than take on the note of malfunctions, I would say this was done deliberately to avoid detection.

I would say the vendors, since they employ people like Jeffrey Dean have gotten quite good at refining their methods. Therefore, to expose the fraud properly you would need to get into their mind.

For another instance, there was a huge discrepancy in Clermont County. If I was a crooked Board Elections manager lobbying for Diebold, I would skirt and delete votes in the less populated urban districts. Why? Because nobody would suspect that democrats were disenfranchised in areas that primarily vote a different way, or third party.

Further, it would appear that these were merely clerical errors...Which I would later blame on my clerk partner, Peggy.

This happened in Florida in 2004 when thousands of fraudulent votes showed up and Constance Kaplan, who was lobbying for ES&S, resigned and left.

I don't believe for a second that was coincidence. If you follow the direct conflict of interest, you will find who is stealing whose votes. As they say follow the money but remember what you are looking for, is numbers of votes being switched. If you were to take one apart, examine the "proprietary code" you are looking for votes that were counted being deleted by "memory card errors" or "switched to the other candidate" by clerical errors. You would be able to tell right away by looking at the audit sheet, compared to your county canvassing board. This would be evident in the programming.

[tommcintyre]

59. kick