Hijinx in Ohio?

[gapower]

Seems like something weird's going on in Cleveland. If you look at the presidential election results for Cuyahoga County, you see that "for municipalities with wards," you are to "find the ballots cast total for each ward and total them." But the total among the wards in each such municipality is a good bit fewer than the total listed for the entire municipality itself.

For example, from the City of Cleveland, there were 214,902 ballots cast for president. But from the separate wards (#1 thru #21) of Cleveland, the total is 165,578. So there are 49,324 "missing" ballots. In Cuyahoga County overall there are 199,288 ballots "missing" by this principal. And they are indeed missing because they are not counted in Ohio's total vote count.

So, OK, some things don't add up, and 200,000 votes are missing from the most Democratic municipalities in the most Democratic county in the most critical state that Kerry lost by 137,000 votes. No biggie. Though mind you, this pattern does not show up in any other county in Ohio.

But here's where it gets weird. Here is the number of "missing" votes for each such municipality in Cuyahoga County:

49324 CLEVELAND
9948 NORTH OLMSTED
9948 WESTLAKE
9948 ROCKY RIVER
9948 BAY VILLAGE
9948 FAIRVIEW PARK
8553 WARRENSVILLE HEIGHTS
8553 BEDFORD
8553 BEDFORD HEIGHTS
7284 PARMA
7284 MIDDLEBURG HEIGHTS
6170 GARFIELD HEIGHTS
6007 CLEVELAND HEIGHTS
6007 EAST CLEVELAND
5724 EUCLID
5724 SOUTH EUCLID
5460 OAKWOOD VIL
5295 BROOK PARK
4744 MAPLE HEIGHTS
4314 LAKEWOOD
4009 NORTH ROYALTON
3146 BEREA
3146 OLMSTED FALLS
2540 BROADVIEW HEIGHTS
2147 SEVEN HILLS
1385 HIGHLAND HEIGHTS
1385 MAYFIELD VIL

In 18 of the 27 affected municipalities, the number of missing votes is exactly the same as the number of missing votes in another municipality.

This just doesn't look right. If I received data like this at my office, I would know that something was wrong. There may be some explanation besides fraud. If so, I hope that explanation is found and made clear to the public during the Ohio recount.

It's just something that needs to be clarified, considering that Walden O'Dell, chief executive of Diebold Inc. and active in the re-election effort of President Bush, wrote to Republicans in a fund-raising letter August 14, 2003 that he is "committed to helping Ohio deliver its electoral votes to the president."

[rockedthevoteinMA]

1. this is strange!

Did you send this to anyone yet??

[gapower]

4. Not yet

No, thought I'd check with you guys to see where it might be of interest... will follow up!

[NamVetsWeeLass]

2. please send this to Bev Harris at BBV...NOW.

Bev is a Member here, and she is the organizer over at BBV, She needs to know stuff like this... I will keep kicking it if need be.

[rockedthevoteinMA]

3. does Hedda or Ida need it too?

Maybe you should rename the thread?

[gapower]

10. What to?

Need a more eye-grabbing title??

[Eloriel]

65. I'll get it to Hedda right away

Thanks. Don't know if she needs it, but I'll make SURE.

Thanks again, folks.

[gapower]

7. Will do.

Sounds good - I'm a bit of a noob here - will check it out.,..

[hippiegranny]

5. well...

the plot thickens, yet again.

[hang a left]

6. Does this remind anyone else of 18,181? n/t

[tandem5]

11. yeah I was just trying to remember that number n/t

[IAMREALITY]

8. Hasn't This Been Explained a Zillion Times on DU already

Or am I missing something here...... I'd swear this is old debunked garbage, but I'm so burnt right now and have seen so much goddamn data I don't know if I'm comin or goin anymore LOL

[NamVetsWeeLass]

13. Thing is, if it hasn't been Bev needs to see it....

I know what you are saying about the numbers though. I hate MATH!!! too many numbers and stuff.....

[Woo]

15.

I'm sure they have a perfectly reasonable excuse for this -- if not, someone is coming up with one as we speak...

I'm tired folks...

But don't stop -- we can't afford to wait until 2006 or 2008

[gapower]

18. Been looking for an answer

Hadn't found an answer yet so this is where I come to ask.

[IAMREALITY]

20. Oh, By The Way

Welcome to DU brother (or sister)

Haven't seen ya yet.

:hi:

[gapower]

24. THanks, yep,

I'm a n00b! :8)

[gapower]

29. Oh, and

brother, with only minor forays into sisterhood! (Friday nites, etc.

[On the Road]

78. Yes, This is the Absentee Issue

the data was poorly presented by the county, but the individual polling places do total correctly to the official vote. (I did this myself from the official canvassing report.)

Large groups of absentee ballots from other areas are added to some wards without noting the fact. That's what the difference is.

[neonplaque]

9. The muni's that match are lumped together...

Perhaps the 'missing' totals come from absentee ballots which weren't allocated to a particular ward?

Anyway, if you are a Clevelander, you know that the matching numbers on the munis is because they are each joined at the hip. ie, HIGHLAND HEIGHTS, MAYFIELD VIL next to each other and probably overlap to some degree with voting wards.

[gapower]

14. Maybe so.

Hm - so they add the same number of absentee ballots to each of the muni's, even though they're the same absentee ballots? I can grasp that if that's so.

[Chili]

22. you are right, those municipalities do share borders

...all of those with the same numbers.

I'm not sure that accounts for missing votes, though, if there are truly missing votes...

[gapower]

30. THey share both

both borders and names, as well. I see how they could be adding on the same list of absentee / or spoiled or whatever ballots. If that's what it is, I'll be reassured.

[CAcyclist]

12. Haven't we seen those numbers - 9948 and 8553 - before?

Wasn't that the same as some of the other places where Bush had "extra" votes? Maybe I'm remembering wrong, though.

[GetTheRightVote]

19. It sounds fishy to me that they would be exact every time

You would think that they would have even one with a add number in cast votes. Sherlock Holmes, calling Sherlock Holmes.

P.S. Sorry about the spelling.

[gapower]

36. Yeah I'm thinking that

it could be that the "sameness" of the discrepant numbers (whatever they represent) is explained by a silly but inconsequential routine of adding the same list of absentee ballots to each of the municipalities in a cluster.

[shraby]

44. I remember seeing those numbers too.

Dammm I can't remember where though.

[Chili]

16. not all Democratic municipalities...

I know these areas, though these are only guesses, I have no statistics (though I could find them)

49324 CLEVELAND - heavily Democratic
9948 NORTH OLMSTED - probably heavily Republican
9948 WESTLAKE - probably heavily Republican
9948 ROCKY RIVER - probably heavily Republican
9948 BAY VILLAGE - - probably VERY heavily Republican, LOL
9948 FAIRVIEW PARK - probably heavily Republican
8553 WARRENSVILLE HEIGHTS - definitely heavily Democrat
8553 BEDFORD - mixed Reps / Dems
8553 BEDFORD HEIGHTS - definitely heavily Democrat
7284 PARMA - definitely heavily Republican
7284 MIDDLEBURG HEIGHTS - probably mixed, more likely leaning Rep
6170 GARFIELD HEIGHTS - mixed Rep / Dem
6007 CLEVELAND HEIGHTS - mixed, but leaning Democrat
6007 EAST CLEVELAND - definitely heavily Democrat
5724 EUCLID - probably mixed right down the middle
5724 SOUTH EUCLID - probably mixed right down the middle
5460 OAKWOOD VIL - definitely heavily Democrat
5295 BROOK PARK - probably heavily Republican
4744 MAPLE HEIGHTS - mixed, probably leans Democrat
4314 LAKEWOOD - probably mostly Democratic, though not sure
4009 NORTH ROYALTON - probably heavily Republican
3146 BEREA - probably heavily Republican
3146 OLMSTED FALLS - probably heavily Republican
2540 BROADVIEW HEIGHTS - probably heavily Republican
2147 SEVEN HILLS - mixed, though probably leaning Republican
1385 HIGHLAND HEIGHTS - mixed, though probably leaning Republican
1385 MAYFIELD VIL - can't tell... probably leans Democratic

Anyone else living in this area, please correct / add to this...

[gapower]

21. Great! Thanks -

I don't know the area at all - I printed out a map today to place where the biggest discrepancies are. Thanks!

[Chili]

25. you're welcome!

...though note that I changed North Olmsted from "mixed" to "probably heavily Republican." LOL, what was I thinking?!

[modman]

69. More dems then you think

Parma and Berea have more Dems than Reps (I live in Parma and lived in Berea.I also was campaign-sign watching,,,2 or 3 - to-1 Dem).

[Chili]

70. hey Modman...!

...you're right - I ran across some other totals tonight that proved I was waaaay underestimating (in my head!) the number of Democrats in University Hts. and Cleveland Hts. (and Beachwood as well). They went overwhelmingly for Kerry.

[newyawker99]

74. Hi modman!!

Welcome to DU!! :toast:

[jhgatiss]

17. They really need to come up with a better way to format their data.

I think this arises from the fact that in congressional election years they allocate the ballots by congressional district rather than precinct. Since the precinct lines don't jive with the congressional district lines, votes get reallocated to another precinct where they belong. (I think.) At least that's how I read the Cuyahoga county's BOE website explanation. Then there's that absentee ballot "parking" business too. I just ignore Cuyahoga county. They seem to have some odd procedures to account for everything. :)

[gapower]

26. Yeah i had to ...

pull the columns apart in Excel to see what they were saying.

[scurvy_n_disastrous]

23. odds of this? !*&!#$?*?

no stats man I, but I sure would like to find out. I venture to say that they are astronomical. 18 out 27 wards share the exact same total!
(2nd post)

[Laura PourMeADrink]

27. Many of these Numbers Divisible by 3

I have definitely seen a post with doubled numbers on DU

But, just playing with some of the numbers - many are divisible by 3. Don't know if that means anything, but seems odd...

8553 / 3 = 2851.00

9948 / 3 = 3316.00

7284 / 3 = 2428.00

5724 / 3 = 1908.00

But..

6007 / 3 = 2003.3333

3146 / 3 = 1048.6666

1385 / 3 = 461.6666

[gapower]

32. And PRIMES!!!

Hey I stayed up late last night looking at whether they were multiples of something - like a program loop that says drop so-and-so many votes and go thru it a random number of times. Close, but no smoking cigar... strictly for debunking....

4*11*19*59
12*829
3*2851
12*607
10*617
6007
12*9*53
6*7*10*13
15*353
2*2372
6*719
19*211
2*11*11*13
20*127
19*113
5*277
2*23*67*67

THat 67-squared, 607, 6007, and 617 had me going though!

[Laura PourMeADrink]

34. How many on ballot for prez?

[Mike Nelson]

28. Is ROSE MARY WOODS typing these figures? (nt)

good grief!

[Carolab]

31. QUESTION:

How many more votes did Bush get in Cleveland than Kerry?

[gapower]

33. Kerry beat Bush

At least in Cuyahoga County overall, Kerry outpolled Bush 2:1. and that's among only the votes that got counted.

[Carolab]

38. Another question

Are these optically scanned ballots or touch screen counties? Do you know?

[gapower]

40. Cuyahoga is old-style

If I understand correctly, there's a handful of counties that went to the touch screen and Cuyahoga is not one of them.

[roseBudd]

49. Cuyahuga like most Ohio Counties was punch card, their card

readers were ES&S. So much attention has been paid to BBV that the fraud potential in Ohio hasn't had much attention. I have asked myself if I was Ken Blackwell and I had to cheat using Plan B because of punch cards, how would I do it? I have been unable to ascertain whether the punch cards themselves may have been manipulated and routed to democratic or republican precincts. In other words were there differing lots of punch cards? Anyone know who sells and who orders the punch cards?

[gapower]

52. There has been mention

of selective availability of working voting machines etc

[Chili]

50. Cuyahoga County: CHADS, baby

...home of the Hanging, Pregnant and Dimpled Chad.

Ah, we cross-posted, Rosebud...!

Don't know the answers to your questions. But I have a question: shouldn't it be easy to recount punchcard ballots? When I voted, I signed a book, was assigned a ballot which had a corresponding number that matched the one next to my name. It should be, theoretically, easy to reconstruct an election in any precinct in Cuyahoga County. Right? I'm probably being naive, LOL.

[roseBudd]

54. Yes punchcards can either be sent back through the counter

and hanging chads do fall off in a second pass and they can be hand recounted. There are BTW 93,000 uncounted punchcards that either show no vote or an overvote for president. These need to be hand counted.

Seriously we need to find out who sells punch cards.

[Chili]

67. Rosebud...

I came across this study – it was made in 2003, and while it was for the state of Washington, it’s got a lot of good info in it:

http://www.lwvwa.org/voting_study/Voting%20Issues%20Study%2003.pdf

I found 3 companies, so far, that still may be manufacturing punch cards and punch card readers (all taken from this study’s appendix): VotoMatic (the infamous Florida punch card company), DataVote and LRC Ballot Reader.

I also found this link: http://www.cs.uiowa.edu/~jones/cards/index.html

If you scroll down to the bottom of that page, there are some links, but I don’t think that’s what we need – looks like those are “punched” cards – what, recyclable cards that can be purchased? Geez.

I can’t find any of these companies online.

[scurvy_n_disastrous]

35. kick

[truehawk]

37. another thread that makes the observation 200,000 votes missing from '00

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=203x63560

[gapower]

39. Yeah, I've seen 2 different figures

Both for 2000 and 2004, I've seen values of approx 650,000 and approx 850,000.

I currently suspect that the 650,000 is what was counted for both years.

[momzno1]

42. to clarify, the message is that there are about

200,000 less votes in 2004 than there were in 2000?

[truehawk]

45. Yes

[gapower]

48. further clarification -

are you asking if that's what I'm asking? If so, no. I suspect the 850,000 number attributed to cuyahoga for 2000 was actually incorrect. The cuyhoga board of elex site says there were only 650,000 ballots cast in 2000 as well.

[mostly_lurking]

53. No, that post is wrong...

I posted this info in the other thread as well. It is a red herring:

Here is the Canvass report for 2000:

http://boe.cuyahogacounty.us/BOE/results/history/2000/1...

Total ballots cast for President = 574,777
(On page 3, near the bottom. You have to add up the totals for the different candidates),

Here is the Canvass report for 2004:

http://boe.cuyahogacounty.us/BOE/results/history/2004/E...

Total ballots cast for President = 652,381
(Search the page for "Kerry" to find the listings for President. Again, you have to total the numbers)

I have no idea where the "218,981" less votes cast in 2004 over 2000 came from. It clearly is not supported in the actual numbers.

[gapower]

55. There are actually 2 sources (both wrong i suspect)

The spreadsheet from That Berkeley Group says that in 2000 there were 323227 votes for Bush and 544640 for Gore = 867,867 votes.

and the same source says there were 215624 for bush and 433262 for kerry = 648,886 in 2004, thus 218,981 fewer in 2004 than in 2000.

But the archive at the Cuyahoga website says the count in 2000 was 192,099 for Bush and 359,913 for Gore so only 552,012 in 2000.

I mentioned a 2nd source for the 850k - will cover that in a separate note.

[mostly_lurking]

57. The 850K+ numbers are certainly suspect

As that would mean a nearly 85% voter turnout in a county of just over 1 million registered voters. Such a massive turnout in a large county is unprecedented in US electoral history.

[gapower]

59. Zactly.

[gapower]

58. The other way you can get approx 850k (also probwrong), this time for 2004

is to sum up what should be the correct numbers from the Cuyahoga website for 2004, using the municipality totals, and no, not incorrectly adding in the redundant ward-level data.

I think at least part of the problem with this option is as you note that the absentee or whatever they are extra 9948, 8853 etc values get added in multiple times.

[TennisGuy2004]

41. Registered Voters vs. People who actually voted?

Maybe the numbers seem skewed because the numbers you're listing are of people who just didn't show up to vote? I don't know, the numbers on that web page are Greek to me.

[jdog]

43. There's another thread going on this...

Cuyahoga County

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=63560&mesg_id=63560

[gapower]

51. Different question but thanks!

[gapower]

47. Their spreadsheet is a mess, but

there are separate columns for registered voters and ballots cast - i'm only looking at the ballots cast column

within ballots cast, 27 of the municipalities have further breakdowns by ward, and these need to be separated out into another column.

the problem is that after you do that, the number of ballots cast in the municipality is always a good bit greater than the sum of its parts.

Actually even that alone is not the problem - the problem is that only the smaller number - the sum of the parts - is being used to calculate the state total.

[Laura PourMeADrink]

46. How many * needed in cities down to science

I remember the talking heads discussing the exact numbers * needed or moreover what Kerry COULDN'T get for * to win PA. Surely they knew this for Cleveland too. Just wishing here, but, say they saw Kerry eclipsing that threshold.....

But, I think there is some stupid explanation for this anomoly. Isn't there something about every other year, they take absentees and spread them and then they don't -- something crazy like that???

Getting back to the divisible by "3". * had three opponents!! Course it would be pretty risky to steal from the 3rd party -- the well's not deep enough there

[jdog]

56. YES!

It is explained at the bottom of this current thread:



http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=63560&mesg_id=63560

[gapower]

60. Thanks all, I think I get it now

I suspect the strange repetition of 9948, 8853, 7284, 6007, 5724, 3146, and 1385 is a bug that makes their printout incorrect, but possibly does not affect their actual total.

Thanks!

[truehawk]

61. they said that that was because of the way that they allocated absentee

Ballots to congressional districts instead of precincts, it still does not explain why the vote total went down from '00 or why precincts that were witnessed to be packed show single digit totals.

[jdog]

62. But isn't that the bad data?

Can we get some good data?

[scurvy_n_disastrous]

63. kick

[DoYouEverWonder]

64. These systems weren't supposed to have bugs

They told us that these systems were totally reliable.

[99Pancakes]

66. Doesn't Kerry have lawyers in Ohio?

Somebody must have a lawyer there we can call.

[Lil]

68. kick

[Woody Box]

71. The absentee votes are counted multiple times

This is the most detailed Cuyahoga results page:

cuyahogacounty.us/boe/results/history/2004/110204_GE_Canvass.TXT

There are 80624 absentee ballots cast for Ohio. They are divided into districts:

Total
7902 ABSENTEE VOTERS - 10-10-21 896
7903 ABSENTEE VOTERS - 10-11-21 1044
7904 ABSENTEE VOTERS - 10-12-25 1426
7905 ABSENTEE VOTERS - 10-13-23 4314
7906 ABSENTEE VOTERS - 10-14-23 5295
7907 ABSENTEE VOTERS - 10-15-23 7284
7908 ABSENTEE VOTERS - 10-16-24 9948
7909 ABSENTEE VOTERS - 10-17-24 2147
7910 ABSENTEE VOTERS - 10-18-24 3146
7911 ABSENTEE VOTERS - 11-07-25 5724
7912 ABSENTEE VOTERS - 11-08-25 8553
7913 ABSENTEE VOTERS - 11-09-21 6007
7914 ABSENTEE VOTERS - 11-10-21 2627
7915 ABSENTEE VOTERS - 11-11-21 2633
7916 ABSENTEE VOTERS - 11-12-25 4744
7917 ABSENTEE VOTERS - 11-13-23 203
7918 ABSENTEE VOTERS - 11-17-24 3313
7919 ABSENTEE VOTERS - 11-98-18 574
7920 ABSENTEE VOTERS - 13-17-24 2540
7921 ABSENTEE VOTERS - 13-18-24 4009
7922 ABSENTEE VOTERS - 14-17-24 2812
7923 ABSENTEE VOTERS - 14-98-18 1385
total 80624


So NORTH OLMSTED, WESTLAKE ,ROCKY RIVER, BAY VILLAGE and FAIRVIEW PARK all get the same amount of votes ( 9948 ) because they belong to House District 16:

7908 ABSENTEE VOTERS - 10-16-24 9948

This allocation of the absentee ballots is very accidentally, however. Some cities get them, some not. Nowhere is explained why they do it like that. I think the huge amount for Cleveland is created by adding several numbers below.

Just a suggestion: is it possible that this massive erroneous overflow of absentee ballots is meant to hide something - irregular absentee votes, for instance?

[kk897]

72. Find Sondra Robinson!

She was the woman in charge of absentee ballots for C. County who quit mysteriously days before the election. I've been trying to track her down.

The absentee ballots were poorly designed so that people could easily mistakenly vote for Bush while meaning to vote Kerry. (Also there were some absentee ballots elsewhere that left Kerry off entirely; still others kept Nader on and were supposed to be covered somehow when it was determined he couldn't be on the ballot--who knows if they got around to that!)

[Lil]

73. Re:This allocation of the absentee ballots

Do you know if absentee ballots have always been divided up this way? Just curious. :-)

[Lil]

75. Do you know if absentee ballots have always been divided up this way?

Repeat - just kick in case someone knows.
Happy Thanksgiving from Australia!

[Lil]

76. kick

[Lil]

77. CUYAHOGA COUNTY: MASSIVE Punchcard machine outages

<http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=84186&mesg_id=84186>

check out this thread

[KaliTracy]

79. Warren County

the place of the LockDown during the vote counts.

Here are results from 2000 and 2004

Warren County:
they had 20,000 more votes for Bush than in 2000 and

7,000 more votes for the Kerry than for Gore

2004 results http://www.co.warren.oh.us/bdelec/voting_results_publis...

2000 results http://www.co.warren.oh.us/bdelec/archive/general_11-00...

What I find interesting is that though more independents ran in 2000 the Democratic percentage went down (2000 -- %27.71 Dem votes) (2004 -- %27.58 votes) while Bush's went up 2.11% -- The independent voters in 2000 cast 1618 votes. The independent votes in 2004 cast 337 votes... unless I'm missing votes (or something else) somewhere, which could be very likely... just seems odd

don't have the precinct breakdown here -- just the overall results

tracy

[Lil]

80. Did anyone find Sondra Robinson? Repeated post 72 below

kk897 (168 posts) Sun Nov-21-04 02:07 PM
Response to Reply #71

72. Find Sondra Robinson!


She was the woman in charge of absentee ballots for C. County who quit mysteriously days before the election. I've been trying to track her down.

The absentee ballots were poorly designed so that people could easily mistakenly vote for Bush while meaning to vote Kerry. (Also there were some absentee ballots elsewhere that left Kerry off entirely; still others kept Nader on and were supposed to be covered somehow when it was determined he couldn't be on the ballot--who knows if they got around to that!)