Need C. Ellen Connaly Kerry out polling #s for these OH counties

[roseBudd]

Auglaize, Brown, Butler, Clermont, Darke, Highland, Mercer, Miami, Putnam, Shelby, Van Wert, and Warren

I am not familiar with all these counties but I find it very interesting that Butler, Warren, Clermont & Hamilton are contiguous.

I know I have seen the exact numbers somewhere and the 4 SW OH counties had the largest out polling differential. I am having no luck with Google today :(

[Botany]

1. I just checked the Sec. Blackwell's site .....

.... they have been changed.

http://www.sos.state.oh.us:80/sos/ElectionsVoter/Results2004.aspx

Counties that I know had Connaly getting more then Kerry now have Kerry getting more
then Connaly. Can you say covering your tracks?

[roseBudd]

2. I can almost remember the exact numbers for the SW OH counties...

One of them was like 12,000 more than Kerry. I do remember of the 4 Hamilton was the lowest.

[Botany]

3. This is big! Smoking Gun Fucking Big!

The #s have been changed ..... they are trying to cover their tracks

Old Results ....

Butler County

Connely 61,559 Kerry 56,234

New Results ...

Connely 14,791 Kerry 56,243

Every County that showed Connaly getting more votes then Kerry now has Kerry getting more
votes then Connaly.

This is big! The man running for Governor of the state of Ohio has changed the returns of the 04
elections so as to hide the fraudulent 04 Presidential vote. Oh Mr. Blackwell you got some 'splaining to do.

[roseBudd]

4. OK I'm not good at math I found this linked from an old DU thread of 11/28

04

http://freepress.org/departments/display/19/2004/944

*Chart of the measurement of margin between Connally and Kerry race

A negative sign immediately below means that the margin for Kerry (in his race) was less than the margin for Connally (in her race) by the indicated amount. All counties where Connally's margin exceeded Kerry's by 2,000 votes or more should be on this list. There are 37 such counties.

County Margin Voting system
Adams -2,299 punch card
Allen -4,579 optical scan
Auglaize -6,592 DRE
Brown -4,363 punch card
Butler -45,457 punch card
Champaign -2,252 punch card
Clermont -22,998 optical scan
Clinton -3,429 punch card
Crawford -2,891 punch card
Darke -6,549 punch card
Defiance -2,050 punch card
Delaware -10,431 punch card
Fairfield -4,104 punch card
Geauga -4,433 optical scan
Greene -9,480 punch card
Hamilton -16,289 punch card
Hancock -5,424 optical scan
Highland -3,588 punch card
Holmes -2,393 punch card
Lawrence -2,567 punch card
Licking -6,265 punch card
Logan -3,610 punch card
Madison -2,394 punch card
Medina -3,768 punch card
Mercer -7,127 punch card
Miami -8,869 optical scan
Morrow -2,057 punch card
Pickaway -2,587 DRE
Preble -3,077 punch card
Putnam -5,327 punch card
Richland -3,464 punch card
Shelby -7,544 punch card
Stark -7,300 punch card
Union -2,339 punch card
Warren -24,785 punch card
Washington -2,203 optical scan
Williams -2,662 punch card

[Botany]

5. i never saw that list ..... This data is from Blackwell's own site

This is big buddy ...... Auglaize showed 7,545 now shows 2,806
Brown showed 7,498 now shows 2,638
Butler showed 61,559 now shows 14,791
Darke showed 9,021 now shows 3,056
Mercer showed 6,919 now shows 2,952
Miami showed 17,770 now shows 4,993
Putnam showed 4,846 now shows 2,596
Shelby showed 8,043 now shows 2,751
Van Wert showed 4,589 now shows 1,583
Warren showed 28,470 now shows 6,604

Somebody is trying to keep the body covered up. I will send Brad Blog this data.

[roseBudd]

6. Here is more from Freepress article from '04

Would the wayback archive help?

From:

http://freepress.org/departments/display/19/2004/1057

In Butler Country, Bush officially was given 109,866 votes. But conservative GOP Chief Justice Moyer was given only 68,407, a negative discrepancy of more than 40,000 votes. Meanwhile, Moyer's opponent, a pro-gay, pro-abortion African-American liberal from Cleveland, was officially credited with 61,559 votes to John Kerry's 56,234.


In Warren County, Bush was credited with 68,035 votes to Kerry’s 26,043 votes. But just as the county's votes were about to be counted after the polls closed on November 2, the Board of Elections claimed a Homeland Security alert authorized them to throw out all Democratic and independent observers, including the media. The vote count was thus conducted entirely by Republicans.

Here Blackwell's certified tally says the slightly funded Connally somehow outpolled Kerry by more than 2,400 votes, nearly 10 percent of his county wide total.

[OnTheOtherHand]

7. are you sure you didn't hit the wrong screen?

http://www.sos.state.oh.us/sos/ElectionsVoter/results2004.aspx?Section=138

has all the "old" numbers for Connally.

To take away well over half of Connally's votes in some counties wouldn't be much of a coverup.

EDIT TO ADD: Looks like you were looking at the (uncontested) primary numbers at
http://www.sos.state.oh.us/sos/ElectionsVoter/results2004.aspx?Section=156

[Botany]

8. I might have ....

opps.

damn!

[roseBudd]

9. I am math challenged what are the actual numbers that connely exceeded

Kerry by Butler, Warren, Clermont & Hamilton?

[OnTheOtherHand]

10. Connally didn't exceed Kerry in Hamilton, although

some find it suspicious that she lost the county by fewer votes than he did. (Connally lost by 7132 votes, Kerry by 22,937.)

As for Butler, Clermont, and Warren, here's what I get (with "Connally votes - Kerry votes = difference" given for each) :

Butler: 61,559 - 56,243 = 5,316
Clermont: 30,068 - 25,887 = 4,181
Warren: 28,470 - 26,044 = 2,426

I honestly don't find these results very surprising in themselves: Chief Justice is a nonpartisan race, and it isn't shocking that Connally would pick up some Bush votes. (Since fewer people vote for Chief Justice, the math suggests that Connally or any other downticket Dem is most likely to draw more votes than Bush in counties that are heavily Republican.)

Of course there are other reasons to be suspicious in these counties, especially Warren. (There are also reasons to be suspicious in Cuyahoga, where Kerry outpolled Connally by almost 50%.)

[roseBudd]

11. Yeah but if you were in SW OH you would realize she had 0 name

recognition. I saw not 1 sign, commercial, campaign literature nothing. Moyer was endorsed and according to a Warren County DUer yards with Bush signs did have Moyer signs. I wanted to vote for Connelly and had been told her nmae but forgot it and decided not to guess and didn't vote for SC.

The other part of the argument is that the downticket race has significant dropoff in completes and SC was last on the ballot. There were about a million less votes cast for SC then P if my memory serves me. And the funny thing is in N OH where everyone had heard her name Kerry got a lot more votes than her as expected because SC is downticket.

One more math question. When a candidate is said to have for example 500 votes they don't deserve (taken from the other candidate) doesn't it mean a doubling of the margin the other candidate lost by, because the numbers I remember for SW OH counties were closer to almost double of what you listed.

Edited to add from table above:
*Chart of the measurement of margin between Connally and Kerry race

A negative sign immediately below means that the margin for Kerry (in his race) was less than the margin for Connally (in her race) by the indicated amount. All counties where Connally's margin exceeded Kerry's by 2,000 votes or more should be on this list. There are 37 such counties.

Warren -24,785 punch card
Clermont -22,998 optical scan
Butler -45,457 punch card
Hamilton -16,289 punch card

I really am math impaired :(

[OnTheOtherHand]

12. I'll respond to the first part later

(except I'll say now that I betcha Moyer's name recognition was pretty darn low too, yard signs or no)

Yes, taking 500 votes from one candidate and giving them to the other would change the margin by 1000 votes -- but I don't see how that would apply here, because for the most part we were comparing Kerry's vote to Connally's. I suppose if you assume that every "extra" vote that Connally has over Kerry was a vote that was stolen from Kerry and given to Bush, then the effect on the margin would be twice as large -- over 20,000 votes instead of over 10,000 votes.

MY TURN TO EDIT TO ADD (grin) : OK, so much for "we." I took your OP as a reference to Phillips -- but yeah, one can compare the margins, and then not only does Hamilton also look anomalous, but the totals are much larger (Connally did a lot better on net margin than Kerry did, because Moyer got many fewer votes than Bush). More soon....

[roseBudd]

14. I love it when the math gifted do the thinking...

That's why I stick to photoshopping Blackwell and Rove...

[OnTheOtherHand]

13. first to reiterate my edit to the other post!

Yes, overall Connally lost by much less in these four counties than Kerry did, largely because Moyer got many fewer votes than Bush did.

At the precinct level, Connally's vote share is approximated by the regression formula 0.554(Kerry%) + 0.20. (We could play with that model, but hey....) That implies, for instance, that in precincts where Kerry only got 20%, Connally averaged around 31%, and where Kerry got 90%, Connally only got around 70%. The break-even is about 45%: in other words, Connally does better on average in percentage terms in the precincts where Kerry got less than 45%.

For absolute margin, we have to take into account that as you pointed out, about 20% of the folks who voted for president didn't vote in the chief justice race. That ends up moving the break-even for absolute margin a bit closer to the 50-50 mark. Lessee, it looks as if Kerry lost 72 counties, of which Connally had a better margin in 54 of them -- and Kerry won 16 counties, in all of which his margin was better than Connally's. Of course he tended to win the large counties.

There are two kinds of people in the world: people who think it makes no sense that Connally would do better than Kerry anywhere, and people who think it does make sense. Shrug. But I'll say again: Connally actually had a more favorable (or less unfavorable) margin than Kerry in most Kerry counties.

[roseBudd]

15. I believe they cheated in counties that are expected to be Bush wins

and that SW OH was ground zero for moving a certain percentage of Kerry's votes to the Bush column because everyone would say what did you expect, it's Butler, Warren, Clermont, Hamilton. But '04 was not '00 and as Mark Crispin Miller noted there was a lot not to like about the incumbent and his approval ratings and the fact that undecided historically break for the challenger. Add it to huge increases in likely democratic voters, higher tunout in the young who did go for Kerry. I think they stole in a lot of places including switching some African American votes as they claimed Bush went from 9% to 16% among Blacks in OH.

Another interesting thing that Warren County managed to pull off was reporting after Cuyahoga by not having enough voting machines unlike the very similar demographic contiguous Butler County and keeping the polls open late so they could be last to report, because waiting to see what Cuyahoga would do was imperative to avoid an automatic recount. I don't know if the cheating was equal to the Kerry Bush margin or just to avoid triggering a recount and allow Bush to declare he had a mandate.

The only way it makes sense to me for Connally to do better than Kerry is if her name was Oprah.

In Kerry Counties in NE Ohio Connelly is the homegirl and Kerry is a stiff white guy.

[OnTheOtherHand]

16. well, many things are possible

Since you mention blacks in Ohio -- unweighted, 26 of 211 black respondents said they voted for Bush (12.3%). 211 blacks across 49 precincts is not going to provide a reliable sample to judge whether the percentage is 9%, 16%, or whatever.

Actually, IMHO it was better for her, in places like Clermont, that her name was Connally rather than Oprah (assuming she had to choose!). It's a good Republican (or conservative Democratic) name. (I imagine that Connally did get some extra love in some NE precincts where they knew her, but even there I don't suppose her name recognition was all that high.)

The automatic recount threshold in Ohio would've been about 28,000, so at least in retrospect, Warren did not make the difference. But I don't think anyone (on any side) has offered a satisfactory explanation of what went down in Warren.

[roseBudd]

17. There are 37 counties out of 88 that Connelly outpolled Kerry

Add them all together, stir in suppression, impossibly high turnout and you have a lot of votes that are suspicous.

In a nutshell though, Warren County lied and they had to lie for a reason. If cops were investigating Warren they would put them all in separate rooms and question them until somebody turned.

[OnTheOtherHand]

18. I will stick with 54 (for margin) or 12 (for raw votes) for now

Maybe you have different data, I don't know. Where did you get 37?

As for the rest, yeah, I would like to know what that terror alert was really about.

[roseBudd]

19. I got it from the Nov 28. 04 article on Fritakis's Free Press

http://freepress.org/departments/display/19/2004/944

*Chart of the measurement of margin between Connally and Kerry race

A negative sign immediately below means that the margin for Kerry (in his race) was less than the margin for Connally (in her race) by the indicated amount. All counties where Connally's margin exceeded Kerry's by 2,000 votes or more should be on this list. There are 37 such counties.

County Margin Voting system
Adams -2,299 punch card
Allen -4,579 optical scan
Auglaize -6,592 DRE
Brown -4,363 punch card
Butler -45,457 punch card
Champaign -2,252 punch card
Clermont -22,998 optical scan
Clinton -3,429 punch card
Crawford -2,891 punch card
Darke -6,549 punch card
Defiance -2,050 punch card
Delaware -10,431 punch card
Fairfield -4,104 punch card
Geauga -4,433 optical scan
Greene -9,480 punch card
Hamilton -16,289 punch card
Hancock -5,424 optical scan
Highland -3,588 punch card
Holmes -2,393 punch card
Lawrence -2,567 punch card
Licking -6,265 punch card
Logan -3,610 punch card
Madison -2,394 punch card
Medina -3,768 punch card
Mercer -7,127 punch card
Miami -8,869 optical scan
Morrow -2,057 punch card
Pickaway -2,587 DRE
Preble -3,077 punch card
Putnam -5,327 punch card
Richland -3,464 punch card
Shelby -7,544 punch card
Stark -7,300 punch card
Union -2,339 punch card
Warren -24,785 punch card
Washington -2,203 optical scan
Williams -2,662 punch card

To put it in a flipped NE vs. SW OH example. Melba Marsh is a relatively well known (name in newspaper stories sometimes) sort of black female democratic judge in Cincinnati. If she were to run for SC justice and be underfunded (no TV ads whatsoever) I would not expect white suburban Bush voters in Cleveland, Toledo & Akron to vote for her in higher numbers than they voted for George Bush. Especially since that far downticket the percentage of completes have drastically declined.

If the chart above is even close to being correct the biggest margins are all contiguous SW OH counties. Butler, Warren, Hamilton & Clermont.

[OnTheOtherHand]

20. OK, it was the 2000-vote criterion that knocked it down to 37 (or 36)

No problem with that -- I just figured that as long as we are talking about it, we shouldn't have competing factoids drifting through the mist. 54 counties where Connally's margin was better, 36 or 37 where it was at least 2000 votes better.

(My disagreements with the chart, based on final returns: Stark doesn't belong on the list, actually Kerry did much better than Connally; Van Wert, which is small, does belong; Washington doesn't quite belong, although Connally did do better than Kerry.)

To the extent that this effect appears in Republican counties (for the reasons I explained above), it can only show up as a big diff in absolute margin in a county with lots of votes. Connally's margin in Mercer County was over 32 points better than Kerry's, her margin in Hamilton was only about 3.3 points better, but Hamilton is so much bigger than Mercer that the absolute difference is much larger there.

But beyond that, most of SW Ohio other than Hamilton (dozens of contiguous counties) broke more sharply for Bush in 2004 than in 2000, while most of NE (and SE) Ohio broke more Dem. So that helps those four SW counties stand out even more. Why SW Ohio did that is open to interpretation.

[roseBudd]

21. Because like Arenebeck and other statisticians assert SW OH was ground

zero for tabulation fraud. Of the 4 contiguous counties in SW OH with the highest connelly anomaly Hamilton is the one with a significant black population, the others are white flight suburbs. Hamilton is actually starting to be an embarrassment for the Hamilton County GOP.

I don't know anything about Mercer but if it's white and trends, red why the hell are Bush voters voting for Connelly in Mercer?

Now if her name was Ellen Winfrey then I would understand.

Either they flipped Kerry votes to Bush and did not think about flipping downticket D votes at the same time or they screwed up and accidentally assigned a significant number of Moyers votes to Connelly.

[OnTheOtherHand]

22. lest we start going in circles --

There may or may not have been tab fraud in SW Ohio (or, if you like, there was -- whatever). I am just reporting that the Connally argument isn't likely to convince many poli sci types.

Yes, Mercer is very white indeed -- 98.4% according to the Census. (BTW an easy way to spot-check stuff like this is to go to census.gov, choose Ohio or whatever under the Find An Area Profile dialogue, and then choose your county. Or you can go via quickfacts.census.gov. Not that there was any reason for you to do this -- just in case someday you want to.) But why should white folks not vote for someone named Connally? You seem to be assuming that they know something, anything about who she is, but what if many of them don't?

The same (general) thing happened in 2000: both the Dem candidates for Supreme Court got more votes than Gore did in Warren, and in Mercer -- and I'm sure in a bunch of other counties. Actually, Alice Resnick got more votes than Gore did statewide. The other Dem candidate for the court, Tim Black, had much the same pattern as Connally: he generally did less well than Gore in Dem counties, better than Gore in Rep counties. The main thing that suggests is that a lot of people have no idea which judicial candidates are Dem or Rep -- and so, the vote shares for judicial candidates vary less than the shares for presidential candidates.

(Of course, it's also possible that there was massive fraud favoring Bush in Ohio in 2000, although Gore actually did better than the pre-election polls predicted. But the comparison to judicial candidates doesn't tell us whether there was fraud.)

[roseBudd]

23. Why would white SW OH Bush voters be so much more likely

to vote for someone they never heard of. Resnick is not comparable she has always had TV ads both for and against her and her name recognition includes a DUI.

Arnebeck is a lawyer who specializes in public interest litigation. I doubt that he was ambulance chancing when he said the Connelly anomaly was prima facie evidence of fraud. Add to that an unsecure warehouse owned by a GOP operative used for ballot storage an unsubstantiated and preplanned terrorist threat that led to independent observers being blocked from watching the tabulation and you have something that does not pass the smell test.

Name recognition is often mentioned as a positive in political campaigns. Connelly had zero name recognition in SW OH.

[OnTheOtherHand]

24. ok, _now_ we can go around in circles (grin)

You are singling out SW Bush voters. But that isn't the point. Bush voters all over the state, in both 2000 and 2004, often voted for Democratic judicial candidates. Why do I believe that they actually did this? Because when I pored over Franklin County election returns back in the early 80s, I saw the same propensity for the judicial races to break out less partisan than the other races -- which makes it possible, even likely, for a Dem judicial candidate to outdraw a Dem presidential candidate in Republican precincts despite the downticket dropoff. Unfortunately, I don't have access to county splits from the 80s or even the 90s, but I wouldn't advise you to bet against me on this.

I have no grievance against Arnebeck -- I certainly didn't accuse him of ambulance chasing. Nor do I have a grievance against Richard Hayes Phillips, who seems to have initiated this line of analysis, and who has done much excellent work on Ohio. But Phillips is a geomorphologist by training, and he missed something fairly obvious here. He thought it was nutty to suppose that some people might vote for both Bush and Connally; he was wrong.

Do not change the subject on yourself. An unsubstantiated terrorist threat doesn't pass the smell test with or without a Connally anomaly. But if the Connally anomaly doesn't count as evidence in its own right, then linking it with the terrorist threat doesn't help the cause. There is neither a terrorist threat nor a Connally anomaly in Cuyahoga, yet I think there is better evidence of actual miscounting there than in SW -- which isn't to say that I rule out miscounting in SW. But evidence has to be assessed, not just tallied.

Yes, of course name recognition is a positive, as long as it is positive name recognition. Beyond that, we can't say much. It's not as if Connally did well in SW Ohio: she got killed in those three solid-R counties (and the dropoff was around the state average). It's just that Kerry did even worse. Using Connally as a baseline for how well Kerry should have done makes no sense. Judicial races are all different, but none of them is anything like a presidential race. Using Gore as a baseline seems more reasonable -- and doing that, the SW counties don't stand out from the pack, except that Kerry did substantially better in Hamilton than Gore did.