Certified Cuyahoga Results spreadsheet (Excel)

[jmknapp]

Here's a spreadsheet which includes the election night unofficial results for Cuyahoga County, the just-certified official result, and a computation of the differences between the two:

http://copperas.com/cuyahoga/cuyahogaofficial.zip (zipped Excel file)

Included are polling places and location (lat/lon) of the precincts.

Joe

[Lil]

1. Valuable, updated EXCEL file for researchers

Also see:

Cuyahoga County voting problems -- map

<http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=201x5919>

[Lil]

2. another kick

[Lil]

3. Kick for OH researchers

[Lil]

4. back to the top for OH researchers

[Lil]

5. kick to the top for researchers

[ccarter84]

7. keep kicking man

I'll watch it for a bit

[ccarter84]

6. precincts i'm particularly fond of

Change in Official results compared to the unofficial count specifically looking at column AO

in Cleveland 7e...Kerry gains 102...Bush/Shrub gains 2
in Cleveland 10L..Kerry gains 263...bush gains 5...odd no?

I'm not too happy with cleveland 11T, where kerry was -141 and bush was -19
but it was more than made up with the changes in absentee voters all the way down in row 1442-1463 (Still column AO)

where kerry gained..3127 more votes while bush only got 1318

anyone else enjoying these numbers and as tired as I am?

(edit)Why does bay village 1-a have 15 more votes cast, but 16 more votes for pres? (columns referenced are:AT & AU)

[Lil]

8. good observation. Only ~10 votes per precinct for each of the 1000 in OH

does the trick, I think.

[Lil]

9. Kick for researchers.

[Lil]

10. big kick

[skids]

11. Full court shot.... he kicks.... n/t

[Lil]

12. These Knapp threads got lost with all the other news. Kick for researcher

Summary:
- registrations up 30% (96,536 to 125,165)
- total turnout up 36% (70,109 to 95,512)
- percent turnout up 4% (72% to 76%)
- Bush winning margin up 12,816 votes (29,176 to 41,992)
- Bush 72% overall
Random gratuitous graph:
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=114120

Ballot Shuffle in Ohio
That is, Ohio law states that the ballot order of candidates be randomized across precincts, so if a card is run through a counting machine for another precinct, the candidate totals will be shuffled.
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=108132

Distribution of Cuyahoga provisional ballots -- map
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=105452

Ballot spoilage in Cuyahoga County
This map shows the Cuyahoga County precincts where spoiled ballots (those deemed to not have a valid vote for president) exceeded 3% of the total ballots:
http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=203&topic_id=92747

[Lil]

13. Keeping this kicked. jmknapp, do you want to repost the above

threads? I had major computer problems for a few days and didn't stay on the ball keeping them kicked and they got behind and can no longer be kicked. And perhaps in the Demopedia also?

[Lil]

14. kick for excellent research

[Lil]

15. kick for the other threads in post 12

[truehawk]

16. kick

[Lil]

17. back to the top for great maps and graphs of many parts of OH

[Lil]

18. kicik so this doesn't get lost

[jmknapp]

19. Thanks lil

I should really put this stuff on one web site.

[Lil]

20. Good idea - but I know research mode and web site developing . . .

. . .mode seem to confict with each other. :-)

Excellent work. Keep it up as much as you are able.

[Lil]

21. Good night from Australia. keep on kickin' n/t

[Lil]

22. 1 more insomniac kick :-)

[Iceburg]

23. Kick ... because I haven't slept since you posted this tasty treat!

[L. Coyote]

24. Kick plus -- Ohio spreadsheet --- Bush shift 2000-2004 charts

Spreadsheeets

[Lil]

25. Thanks for these, L. Coyote

[L. Coyote]

29. Ohio spreadsheet just updated, more charts and summary

plus some corrections. If you downloaded before, get the update, 541 Kb.



This is the statistical summary for percentage of non-votes per County, an attempt to quantify spoilage by machine type. The mean is 1 percent lower for E-Voting.

Okay, so punch card voting only deprives you of one percent of your vote!! What's one percent in an American election?

Coshocton County, with 7.8% non-voting for Pres. is skewing the standard deviation of Op-Scan counties. What's up with that number??

It seems that for every variable, voting machines make a big difference. Aren't they supposed to be independent variables?

[BlueDog2u]

36. Voting Machines Should be Indpendent variables

It seems that for every variable, voting machines make a big difference. Aren't they supposed to be independent variables?

I think we should keep saying that.

[L. Coyote]

45. Not in New Mexico .. check out this good thread ..

http://www.democraticunderground.com/discuss/duboard.php?az=show_topic&forum=168&topic_id=328#354

[dubyaD40web]

26. Kick

[59sunburst]

27. Clues as to how Cuyahoga may have been hacked

All you guys who are looking at Cuyahoga please review this Excel file:

http://sunburst59.home.att.net/cuyahoga_cookbook.zip

I've compared precincts from 2000, the first 2004 tally and the final 2004 tally. I noticed that 47 precincts have the exact number of Bush votes from 2000 to 2004 (first and/or final tallies).

This seemed like a high number. I wondered if Bush's 2000 numbers were somehow used as a benchmark for each precinct in 2004. Through much trial and error, I was able to reverse engineer a formula that may approximate how the vote was hacked.

Please check out the tab named Zero.

I was able to predict the exact Bush and Kerry results (initial and final) for the following precincts:

1006 BROOKLYN F
1705 CLEVELAND 3E
2115 CLEVELAND 7O
2502 CLEVELAND 11B
4113 GARFIELD HTS 1-C
4901 LINNDALE A

The calculation uses:

A) Total independent votes
B) Total of all counted ballots
C) Total of all cast ballots
D) Total 2000 Bush votes

Using this calculation, I was able to predict Kerry's vote total within +-5% for over 600 precincts. This is on the "Kerry 5% tab".

It's not perfect but I think it shows that that Bush's 2000 numbers were used to alter the numbers.

I welcome all comments and I invite you to improve upon it in any way.

[Lil]

28. Thanks for the inf. I have no expertise in that area, but it does look. .

. . .interesting, doesn't it. :-)

[59sunburst]

30. More than coincidence

What struck me was how the formula worked for those precincts on both counts and yet Bush's count has changed from 2000.

The formula works roughly like this (for the final 2004 tally):

1) Determine the number of independent votes. A factor may have first been applied to these votes to lower them.

Column T

2) Determine Bush's 2000 vote.

Column F

3 Determine ballots cast 2004.

Column C

4)Determine total votes counted 2004.

Column E

5)Determine percentage of independent votes against votes counted and round down.

=ROUNDDOWN(T/E,3)

Column Z

6)Determine percentage of votes counted against ballots cast and round down.

=ROUNDDOWN(E/C,4)

Column X

7) Determine Independent vote offset and round up. I label it Kerry factor but it's really an offset for the independent votes.

=ROUNDUP((0.0107+Z)*X,3)

Column V

8) Determine Kerry votes and round down.

=ROUNDDOWN((1-ROUNDDOWN(F/E,2)-V)*E,0)

Column Q

9) Determine Bush votes. It's just what's left over.

=E-(Q+T)

Again, I'm not saying that this is the exact formula, but I do think it shows that the 2000 Bush numbers were somehow used.

[newyawker99]

31. Hi 59sunburst!!

Welcome to DU!! :toast:

[59sunburst]

35. Thank You

I'm working on some refinements to this calc. I'll post it when it's ready. There also seems to be a correlation with Bush's 2000 percentage of counted ballots. It might be interesting to try it against any other counties with precinct level data available from 2000 and 2004.

[BlueDog2u]

37. Interesting/kick!

I'm not qualified to address your methodology, but thanks for the intriguing analysis.

[K3RRY1]

32. N00B Kick !

:-)

[Lil]

33. kick for all the good posts and links in this thread n/t

[Lil]

34. kick to the top

[sundancekid]

38. kick - interesting number crunching thread and hacking to the dark side

[dlaliberte]

39. Correlate precinct size with differences

I posted a report from the Cuyahoga county counters at:
http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=203&topic_id=175361&mesg_id=175434

One of the relevant facts is that the 3% of precincts that were hand counted all had 550 votes or more. So please look in the spreadsheets to see how those precincts correlate with the differences.

[Lil]

40. another kick to the top

[Lil]

41. Cuyahoga kick

[Lil]

42. kick

[L. Coyote]

43. Breakdown of locations and number of candidate sequences

Precincts w/ # Ballots @ # Locations	

10	5	1
9	4	1
7	5	1
7	3	1
6	5	2
6	4	3
6	3	1
5	5	4
5	4	22
5	3	6
4	4	28
4	3	26
4	2	6
3	3	106
3	2	43
3	1	2
2	2	176
2	1	18
1	1	134

Notice the number of locations that have less sequences than precincts. This would allow easier manipulation of which candidate does not lose votes to cross-precinct voting.

[Lil]

46. L. Coyote, Help. I don't think I understand this. Thanks.

For instance:

did 10 precincts with only one voting location each have ballots with 5 different sequences?

did 1 precincint with 134 locations . . .?

[StephanieMarie]

44. kick n/t

[Moderator]

47. Locking.....

Excessive kicking is not allowed.


DU Moderator