Punch cards and Precinct codes, a tale of 2 states

[roseBudd]

Back in November when I was trying to figure out how one would cover one's ass if they committed tabulation fraud in a punch card county I corresponded with a man who had a web site extolling the virtue of punch cards over DREs. He was a citizen pre-election tabulation testing observer in Indiana. I mainly asked him how easy it would be to do (very) and if it was done how the perp might go about swapping new ballots in event of a recount. He stated that the ballots would be tricky because of prepunched precinct codes on each ballot. Based on that conversation I assumed that the punch cards in Ohio were punched with the precinct code.

I mean why wouldn't they be? Isn't that what the punchs are for? They mean something and the computer does something with that information.

In Ohio the precinct codes are printed on the ballot and a header card has to be read everytime a new precinct is tabulated. Which introduces the chance for human error or malfeasance. In addition Ohio has ballot rotation so a simple wrong precinct tabulation will result punchs being assigned to the wrong candidate.

I was reminded of this when I voted in a mayoral primary and I asked the poll workers had it always been like that in Ohio? Was it like that under SoS Sherrod Brown, or was this instituted under Blackwell. They did not know but immediately understood how easy it would be to switch votes by sending the ballots through as if they were from another precinct.

[FogerRox]

1. Rosebud-- sometimes something so simple

can be sooooo big

[NoBushSpokenHere]

2. Also

It would be easy to rig an election using a punch card tabulator if you knew the BOE's would lean toward picking precincts that were close to the 3% --- rig the precincts that weren't close to 3% so that if a recount were held, and the Board of Elections thought they were to pick a precinct to recount that closely matched the 3%, it could conceivably go undetected. Or, if the Boards were instructed to pick precincts where the vote totals for both candidates were closely matched......the ones rigged wouldn't be closely matched so there again, would go undetected.

Yes, I realize the law is for random selection, but as we all know, random selection did not happen in many Ohio counties. I wonder how so many "made the same errors of violating the law?"

[roseBudd]

3. I was initially trying to figure out what Warren County might be doing

with the phony terror alert they used to prevent observation of the tabulation. They did finally let the one Kerry lawyer up, but he had to split his time between 2 rooms.

Warren County was one of three counties with the largest C. Ellen Connaly anomaly.

Find out when or why Ohio ballots don't have precinct codes punched and you may figure out how and in which precincts "mistabulation" could have been used to steal Ohio.

[Time for change]

4. I don't know how relevant this is to what you're saying, but

Here is a study that I did back in December, which showed that Kerry did worse relative to his 2000 performance in counties that used Triad punch cards than with the use of any other voting technology:

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=195808

[Bill Bored]

5. Rosebud, IMO this is the most under-reported and perhaps the most

important aspect of the Ohio voting "system!"

I really wish it were more widely known.

But I still have one important question which I hope you can answer:

Would this be detected by a manual recount or not?

You say "In Ohio the precinct codes are printed on the ballot..."
Does this mean a human could read the precinct codes on every ballot during a manual recount, but the machines can't? Or did you mean to say they are NOT printed on the ballot and the header card is the only place where the precinct code can be found?

Thanks again for bringing this up because I think no conclusive analysis of what happened in Ohio in 2004 and/or 2000 (when Blackwell was also SoS) can be undertaken without this information.

[kiwi_expat]

6. I think that the precinct-codes are printed on the header cards only.

Liam_laddie would know. I've looked through all of my messages from liam. This is the most relevant comment, "Absentees have a rubber-stamp on back-side to identify the precinct. Regular,do not have this...". I assume that means that regular cast-at-poll cards have NOTHING on the back to identify the precinct.

If that is true, a manual recount would not catch mis-placed precinct header cards.

[Bill Bored]

7. Not to mention votes!

"If that is true, a manual recount would not catch mis-placed precinct header cards."

It's not the header cards I'm worried about; it's the ballots!

[kiwi_expat]

8. I am sorry I did not spell that out more clearly.

I'll try again:
If the ballots do not have the precinct IDs shown on the back, a mis-placed header card would not be caught in a manual recount AND THUS THE RELATED (MIS-TABULATED) VOTES WOULD NOT BE CORRECTED.

I have PMd liam asking him to verify that the precinct IDs are in no way indicated on the back of Ohio regular cast-at-poll ballot punched cards.

If he says that is true, a reference to this thread needs to be included in the "brainstorming" thread, under String #3 (evidence suggest that loss of these votes cost Kerry the Electoral College).

Good work, Rosebud and Bill !

[Bill Bored]

11. You go Kiwi!

And also include it in this thread about Votewatch (ESI)!
<http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x395903>

This is something that needs a lot more "hammering."
(Maybe we can get Tom DeLay to help?) ;)

[Blue Shark]

10. Remember...the Republicans fought...

...tooth and nail to make sure the was NO recount in Ohio. Terrific way to show the world who actually won.

[kiwi_expat]

12. I FINALLY FOUND THE ANSWER !

When looking through some old e-mails from liam_laddie I found this:
"Cuyahoga County uses punch cards. The city, ward and precinct is printed on the back of the ballot card for the ballots at a polling location. For absentee ballots it is written by hand on the back of the ballot."

So this means that, at least in Cuyahoga county, a human could read the precinct codes on every ballot during a manual recount, but the machines can't.

Cuyahoga is the county where Iceberg did his/her cross-precinct error research. Significant cross-precinct error was very likely to have occurred there.

Please lets try to find someone who has an "in" with Kucinich and can persuade him to donate a worker to audit the Cuyahoga ballots under the Ohio Sunshine laws.

Cheers!

[Bill Bored]

13. Yippee!

But how come you didn't just ASK Liam?

This is powerful evidence for the motivation to avoid a hand count!

So at least parts of Ohio are verifiable after all!

Count those cards!

On Edit:

But.....suppose the ballots were switched BEFORE they were cast?

I'm still trying to wrap my brain around this but I think that this only solves the counting problem -- not the casting problem.

Suppose that when the ballots were handed out, they were taken from the wrong stacks? The voters would end up voting on the wrong ballots, but they could still be counted on the right machines. Now also suppose you had a poll worker who switched the ballot based on the codes?

So you have dishonest/poorly trained poll worker A handing out the ballots, but honest poll worker B who puts the ballot in the wrong pile for counting based on the precinct code.

Any thoughts? Could this have happened?

[kiwi_expat]

14. I did ask him about Hamilton County. He couldn't remember.

(He has covered a lot of detail in his investigations. I'm glad I kept his e-mails!)

I agree, this only solves the counting problem. There are several ways that the cards could have been mis-PUNCHED to the wrong precinct. Some might even be innocent errors.

I remember reading somewhere that an election worker kindly told a voter that she could use ANY available voting machine to punch her card. (This was in a multi-precinct polling place.)

Sorry!

[Bill Bored]

16. Oh yes, step right up, use ANY machine, ANY machine at all!

If it's a Kerry stronghold, he'll get plenty of other votes, he won't need yours!

And to make it plausibly deniable, all they had to do was TRAIN the poll workers to do it this way. They don't need to know about anything as technical as Precinct Codes. And if anyone asks questions, it's the poll workers' word against the BoE's. It becomes a "he said she said" argument.

"Don't you remember we told you not to send those poor voters to just ANY machine?"

Or how about just pre-mixing the ballots before they're handed out and neglecting to tell the poll workers to check the codes?

[kiwi_expat]

15. Iceberg's suggestions for auditing for mis-PUNCHED cards (wrong prct)

Please see
http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x337293#354185

[Bill Bored]

17. Yeah the Burg was the Best!

I hope her data got into the hands of some serious people on our side of that border! I have a feeling that if Shrub takes a fall, we may never know all that contributed to his undoing.

[roseBudd]

21. Exactly BUT a ballot could be purposely mistabulated because the

card reader has no idea what precinct the ballot belongs to due to the fact that the precinct code is printed rather than punched and card reader read punchs not ink.

[kiwi_expat]

23. My point was that tab error could be caught - in a manual recount.

Bill had asked whether that was possible or not.

I agree that the lack of a punched precinct-ID on each ballot is a totally unacceptable situation!!

[roseBudd]

24. not so much an unacceptable situation but a situation that causes you

to ask why. Why not do it the easy way, the way the technology is supposed to work and instead choosing a manual way.

To me, printed rather than punched ballots in one state and punched ballots in a contiguous state is very suspicious.

The question is why, when and who decided to have the ballots printed rather than punched. Why was Dayton Legal Blank chosen to be the only vendor in Ohio for ballots? Did Blackwell choose them?

[mod mom]

9. Excellent theory Rosebud, passed on to Ohio election reformers!

[eridani]

18. Ohhhh, SHIT!!!!!!!

http://www.cleveland.com/news/plaindealer/index.ssf?/base/news/112945506310060.xml&coll=2

Ohio is replacing all of its punch-card and lever machines with more-modern technology in the wake of legislation passed after the controversial 2000 presidential election. A handful of counties will be ready this fall, but most will debut their new technology -- mostly made by Diebold -- in 2006.

Further fueling the burgeoning "voter advocacy" or "voter rights" movement is that many of the conspiracy theories have a tantalizing, if partial, basis in fact.

[roseBudd]

20. It is a shame we spend so much ink in Diebold, all PC tabulated ballots

are vulnerable to mistabulation fraud. The assumption is that the existence of a punch card as evidence is a powerful deterrent to fraud. However the odds of a hand recount uncovering fraud are low, because of what it takes to trigger a hand recount.

Cuyahoga definitely had poll workers who told people they could cast ballots on any machine.

Hamilton had 500 ballots invalidated for a voter voting in the wrong precinct in a multiprecinct polling place.

We still need to answer the question WHY IN THE HELL would a punch card NOT have the precinct code punched on the ballot. Indiana does, Ohio doesn't. If the precinct code is punched, the tabulation program would automatically interpret the ballot correctly. The precincts would not have to be preceded by mannually placed precinct header cards.

Does anyone know how to contact Sherrod Brown and ask him? This is something Andy, bless his heart, was going to do.

[roseBudd]

22. What the IN punch card voting expert told me about punch cards...

He was a citizen volunteer observer who monitored and witnessed election procedures including preelection tabulation tests.

I asked him if tabulation fraud had occured what it would take to cull and replace ballots to cover up the discrepency if there were a hand recount. I assumed a machine recount would just retabulate fraudulently.

>>In response to your question about whether the
>>vendor would have to be in on it to achieve fraud,
>>the answer is that only a single programmer with
>>access to the tabulation equipment would have to be
>>in on it. If you gave me access to the tabulation
>>program and some time to analyze it, I could easily
>>produce a "switch" triggered by an unusual vote
>>combination, a time-based trigger, a threshold of
>>votes cast, or any combination of the above.
>>
>>In regard to changing punch cards, this would be
>>slightly more complicated, but possible. One issue
>>that makes it more complicated is that the deck of
>>"blank" cards used in each precinct is pre-punched
>>to identify the precinct, so you'd have to get a
>>correctly pre-punched deck that matches. There's
>>also the issue that (assuming procedures are
>>followed), the cards are always within sight of a
>>representative of both major parties and sealed when
>>they are not, so you'd have to either find a
>>"traitor" to work with or thwart the sealing
>>system. In my opinion it would be simpler to modify
>>the program. I know that in Wayne county IN, before
>>I modified the local procedures, the computer and
>>disks containing the program were not kept under
>>seal following the public tests. It is my guess that
>>many localities are lax in this regard.
>>

[eridani]

19. A blast from the past

http://www.ohio.com/mld/ohio/news/11352898.htm
http://www.dailykos.com/story/2005/4/10/14949/0380

Erie County Prosecutor Kevin J. Baxter is investigating whether the Cuyahoga County Board of Elections broke the law in its recount of ballots from the November presidential election.

The probe stems from two requests written to Mason: one from minor-party presidential candidates David Cobb and Michael Badnarik, and another from entrepreneurial consultant Edward Michael Caner.

Dalton said Mason turned over the papers March 2.

The complaints allege that the board violated state law because the precincts it recounted were neither randomly selected nor was the opening of ballots properly witnessed.

In addition, Cobb and Badnarik allege that there were problems with the board's ballot-transfer cases, which can reveal whether the precinct used the ballots assigned to it or whether ballots from other precincts were used.

Finally, they contend that the county's vote-tabulation machines were used improperly and that discrepancies exist between the certified recount and the certified original vote.

All this was done to cover up problems in the November vote and ensure that no hand-recount would have to be done around the county, the letter from Cobb and Badnarik alleges.In addition, the candidates' letter contends that the way the precincts were chosen seems "to be of a special sort: those in which (U.S. Sen. John) Kerry received either his largest or second largest number of votes in the ward. This meant that precincts in which (President) Bush received an unusually high number of votes could not be examined, nor could the precincts in which the third-party candidates received unusually high vote totals.''

The letter said there is no way this phenomenon happened at random.

More allegations

The letter alleges that Maiden admitted in a Dec. 22 meeting that "ballots in selected precincts had been opened without the presence of witnesses and had been sorted and hand counted in advance of the original recount'' -- setting up a test-run to assure that the recount would comport closely with the original count so that a full hand count wouldn't have to be conducted.

As for the transfer-case problem, the letter alleges that on Dec. 17, a number of precincts were found to have had problems -- namely some ballots assigned to one precinct were used in another, or too many or too few ballots were used.

The letter suggested that "the (election) staff had been assigned to clean-up the tell-tale evidence of election irregularities within the cases.''

Kerger said Thursday that Cuyahoga County was the only county to receive a letter like the one he referred to Mason's office. He said generally he understands that county boards of elections, mostly made up of volunteers, aren't going to run perfect elections. "If we hold the Super Bowl every four years, we wouldn't expect the referees to be perfect,'' he said.

However, what he found in Cuyahoga County was different. There "it seemed to be more than just a mistake,'' he said.