OHIO: 2004 vs. 2000 Judicial vs. Presidential Votes -- WTF???

[Bill Bored]

OK, we all know about C. Ellen Connolly, the African-American Democratic candidate for Chief Justice of the Ohio Supreme Ct. who lost her race by a narrow margin and outpolled John Kerry in 12 counties right?

Well, you ain't seen nothing yet -- unless you were paying attention to Ohio in 2000.

In 2000, there were not just one, but two judicial races in which the losing candidates outpolled President-Elect Al Gore in many more counties than Connolly outpolled Kerry and by many more votes too.

Here are the numbers:

2004:

Kerry-Connolly
Counties where the losing judge outpolled Kerry: 12
Vote difference: -19,621
Presidential Race Margin: 51-49
Kerry Vote: 2,741,167
Judicial Race Margin: 53-47
Connolly Vote: 2,073,886
Statewide Vote Diff: 667,281

2000:

Gore-Black
Counties where the losing judge outpolled Gore: 40
Vote difference: -42,761
Presidential Race Margin: 52-48
Gore Vote: 2,186,190
Judicial Race Margin: 52-48
Black Vote: 1,869,060
Statewide Vote Diff: 317,130

Gore-O'Donnell
Counties where the losing judge outpolled Gore: 45
Vote difference: -123,839
Presidnetial Race Margin: 52-48
Gore Vote: 2,186,190
Judicial Race Margin: 57-43
O'Donnell Vote: 1,740,516
Statewide Vote Diff: 445,674

All 12 counties in which Connolly got more votes than Kerry in 2004 were among those in which both Black and O'Donnell outpolled Gore in 2000. However, in 2000 there were many more such counties!

I'm sure there are a number of perfectly valid poli-sci-based alternative explanations to election fraud for this, and I'd be happy to hear all of them. Surely Bush couldn't have stolen 2 elections in a row in the Buckeye state. After all, it was FLORIDA, the Sunshine State, where they had all those butterfly ballots, hanging chads and all that stuff in 2000, right?

And we all know that because the Ohio results of 2004 match the ones from 2000, that 2004 just had to be on the up and up, right?

And just because the League of Women Voters is suing the state for election dysfunction going back 30 years, doesn't mean things weren't on the up and up in 2000, does it?

Of course not! They just like to vote for judges during security lockdowns in Warren county, that's all.

To really test this, we should look at other states and counties to see how frequently losing judges are supposed to outpoll losing presidential candidates in close races. Any volunteers?

[Old and In the Way]

1. Make a good poli-sci paper this fall.

I'll go out on a limb and say that this is statistically, very odd. Presidential races are always the #1 vote getter....at least it used to be when we had paper ballots.

[Botany]

2. my state is so dirty

:kick:

[pf99]

5. It's not your state it's Ken Blackwell and the corrupt Republicans

Ohio is the state of my birth and the people of Ohio are some of the best in the country!

Its time to clean up the mess and vote out those corrupt Rebulicans.

[Son of California]

3. I wish there was a way to crunch these numbers

In a more idiot-friendly fashion.
This feels to me, and I think anyone with the brain to do a little math, something close to a smoking gun...

[yowzayowzayowza]

4. I've noticed weird patterns like that before:

Election nite 2004 Florida - http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x1296683#1297983

[roseBudd]

6. We need to ger Sherrod Brown involved. He used to be OH SoS

He may be able to shed some light on current practices.

I want to know why punch cards don't have the precinct codes punched to ensure that they are tabulated with the right tabulation program. In Indiana the punch cards have precinct codes punched. Why not Ohio?

[Bill Bored]

7. Have you emailed Matt Taibbi about this?

Also, are there undervote numbers by county anywhere?
I'd like to see if there were more undervotes in the counties where the judges did so well compared to Gore and Kerry.

[Bill Bored]

8. Poli-Sci-Based Explanation

The only one I've heard so far is that judges' party IDs are not listed on the ballot, so voters can't tell if they're voting for Dems or Repubs. They could end up voting Repub for President and Dem for Judge, even if this isn't their intent.

If we can find some other states with the same setup and similar margins and see how well their judges do compared to presidential candidates, it might provide some insight.

Also, the numbers in each county need to be looked at in more detail.

[FogerRox]

9. A big freakin poly sci kick for Bill and this Thread

I just reco'ed fot GT I suggest we all do it

[minvis]

14. A list of states that have elections for appellate judges

Here's a link to a website that lists which states have either partisan or non-partisan elections for appellate judges.

http://www.justiceatstake.org/contentViewer.asp?breadcrumb=5,128

Partisan Elections
The following states hold partisan elections for at least some level of appellate judges:
Alabama
Illinois
Louisiana
New Mexico
New York
Pennsylvania
Texas
West Virginia

Nonpartisan Elections
The following states hold nonpartisan elections for at least some level of appellate
judges:
Arkansas
Georgia
Idaho
Kentucky
Michigan (ii)
Minnesota
Mississippi
Montana (iii)
Nevada
North Carolina (iv)
North Dakota
Ohio(v)
Oregon
Washington
Wisconsin

Retention Elections
The following states hold retention elections at some point to retain at least some level of
appellate judges. Retention elections may follow appointment or partisan election:

Alaska
Arizona
California
Colorado
Florida
Illinois
Indiana
Iowa
Kansas
Maryland
Missouri
Nebraska
New Mexico
Oklahoma
Pennsylvania
South Dakota
Tennessee
Utah
Wyoming

For detailed information, please see:
American Judicature Society
http://www.ajs.org/select11.html

ii While technically nonpartisan, Michigan judicial candidates are nominated at party conventions and often
run with party endorsements.
iii If an incumbent judge is unopposed for reelection, a retention election is held for that judge.
iv In 2002, North Carolina passed a law providing optional full public financing for appellate-level judicial
elections. This law also changed these judicial elections from partisan to nonpartisan. The change will go
into effect in 2004.
v Legally nonpartisan, Ohio judicial candidates run in partisan primary elections and often run with party
endorsements.

As you can see, Ohio is listed as non-partisan but they do run in partisan primary elections and often are endorsed by one of the parties.

If anyone is in Michigan and can look at those numbers, that might be a good case study. Michigan's presidential race was close, it is a neighboring state to Ohio and although non-partisan in judicial races (like Ohio), the judicial candidates are endorsed by the parties.

[OnTheOtherHand]

16. thanks, minvis

That's how I remember it from the 80s -- as far as the party operatives were concerned, the judicial races absolutely were partisan, but they appear as nonpartisan on the ballot. And since they don't have big campaign budgets, and most voters aren't paying attention to them, accidental crossover voting is very easy.

"Brown" was a good Democratic name in Ohio in the 80s (I can't for the life of me remember why -- I think a popular governor before I even moved there at age 7), and the Dems would gripe about Republicans fielding Browns for judicial races just to get some votes from confused Democrats.

What would really be startling, a priori, is if the total vote for all candidates in an Ohio judicial race were higher than (or even close to) the total for the presidential race.

[eridani]

10. Results for Washington State

Washington State—mostly optical scan, DREs in Snohomish and Yakima (part of CD 4) Counties. The contested State Supreme Court judicial races are in ballot order here, and show the expected lower vote totals for races further down-ballot in the most conservative CD, the most liberal CD and the problematic Snohomish County.

Maverick Richard Sanders has very high name recognition, and so has a decisive win everywhere over a less well-known opponent. Becker wins handily in liberal CD 7, loses in conservative CD 4, and narrowly loses in swing county Snohomish. These results are what you would expect.


CD 4 WA state (most conservative CD)

Kerry – 90026
Bush -- 160244
Nader -- 1771
Total – 252041

Gregoire (D) – 83304
Rossi (R) -- 162706
Bennett (L) -- 4601
Total -- 250611

Mary Becker (more liberal) -- 80085
James Johnson (more conservative) 126944
Total -- 207029

Richard Sanders (maverick Libertarian) -- 123080
Terry Sebring (lawnorder Dem) -- 73235
Total -- 196315

CD 7 WA state (most liberal CD)

Kerry – 283865
Bush – 68600
Nader -- 3010
Total – 355175

Gregoire (D) – 253074
Rossi (R) -- 87214
Bennett (L) -- 8610
Total -- 348898

Mary Becker (more liberal) -- 177927
James Johnson (more conservative) 95145
Total -- 273072

Richard Sanders (maverick Libertarian) --155627
Terry Sebring (lawnorder Dem) -- --102453
Total -- 258080


Snohomish County (weird DREs) WA state

Kerry – 156468
Bush -- 134317
Nader -- 2268
Total –293053

Gregoire (D) –139284
Rossi (R) --145404
Bennett (L) -- 10634
Total -- 295322

How odd! CDs 4 and 7 have fewer total votes for governor than for president, as you might expect. Snohomish, with the weird difference between DRE and mail-in optical scan totals (per Land Shark lawsuit) has a higher total for the gubernatorial race.

Mary Becker (more liberal) -- 112529
James Johnson (more conservative) 129828
Total -- 242357

Richard Sanders (maverick Libertarian) -- 141457
Terry Sebring (lawnorder Dem) -- 91586
Total -- 233043

In a word, this is nothing like Ohio. It's more what you would expect of a state that had two machine counts and one handcount of the gubernatorial race come out within 0.01% of each other. (Yakima and Snohomish DRE votes were not recounted, of course.)

[eridani]

11. Anybody else? n/t

Ohio and Florida are weird.

[Bill Bored]

13. Well, the losing judges in CD7 did get more votes than Bush.

This was a Kerry stronghold.

In Ohio, 2004 and 2000, the counties in which the judges outpolled Kerry were all Bush strongholds. So perhaps it's not that unusual.

Are the judges' parties shown on the ballot in WA?

[eridani]

15. No, judges' parties are not shown

And now I get what you are asking about. Judicial races probably tend to be split more evenly because people are much less informed about them. Looks like this is probably not indirect evidence of fraud. What I thought people were saying is that vote totals for down-ballot races were higher than those for the presidential race. Wasn't there something about ballots with votes for every position except president? Or am I thinking of something else?

[Bill Bored]

19. Well, there were 93,000 undervotes, but they do not seem to be

correlated with the counties in which the judges outpolled Kerry.
But that would have been a good theory.
Also, I found no counties where the vote totals for both judges were higher than that for Pres (which is another way of saying the same thing).

So far though we've only found one similar case in WA.

It might be interesting to see if it's as common as in Ohio.

I agree your Snohomish races are suspect. In fact, by definition, you should see quite a few undervotes for Pres. if there are more votes for Gov. there.

[autorank]

12. Bored Rules! This is realy first rate. Simple, to the point.

I appologize for not recommending it (since I missed the 24 hour margin).

Hey Bill, there are just some places where the state supreme court judge is part of the fabric of every day life.

:rofl:

[Time for change]

17. I used to think that the Ohio judicial races held clues to what happened

in Ohio.

But after OTOH pointed out to me that they are probably not listed as partisan races on the ballot (so that ticket splitting wouldn't be necessary to give more votes to Connally than to Kerry), and nobody has said otherwise, I have come to believe that this is not a crucial issue.

I do believe that Kerry was screwed out of a lot of votes in southwestern Ohio (or rather that phantom votes were added to Bush's total), but not mainly because of the Connally anomaly. I think that the fact that the same thing happened with Gore in 2000 makes the fact that it happened to Kerry less significant.

I think that one thing that might shed some light on this would be pre-election polls, especially if they were available by region.

[OnTheOtherHand]

18. pre-election poll by region

Your wish is my command?!

I won't carry water for the Dispatch poll, but here is their survey reported 10/31, in which they had the race dead even --

http://www.dispatch.com/election/election-president.php?story=dispatch/2004/10/31/20041031-A3-00.html

with regional splits. I don't know whether there was a later one. I also don't know whether any of the other Ohio surveys reported regional splits.

Here is the narrative, which emphasizes that Kerry has the momentum (although on statistical grounds that argument might have been a bit flimsy) --

http://www.dispatch.com/election/election-president.php?story=dispatch/2004/10/31/20041031-A1-00.html

(N.B.: DU truncates the display of the URLs, but they actually are different!)

[Time for change]

20. Thank you OTOH, and sorry for the misunderstanding

but I was talking about polls for the Connally-Moyers race.

So much has been made out of her lack of name recognition, yet she ran a respectable race against Moyers state-wide. Apparently many people don't believe that she could have ran that well, and that her strong showing has something to do with the fraud that gave Bush the "win" in Ohio.

I thought that a pre-election poll on this might shed some light on the matter. Did Connally really do so much better than expected? And if so, what did that have to do with the state-wide fraud that cheated Kerry out of Ohio? I can't answer any of those questions.

[minvis]

21. The second link shows the poll for Connolly v. Moyer

And in perhaps the poll’s biggest surprise, Ohio Chief Justice Thomas J. Moyer is struggling to win re-election. He has a 2-point lead — exactly the poll’s margin of error — over Democrat C. Ellen Connally, a former Municipal Court judge in Cleveland.

The other two Republicans seeking seats on the high court — incumbent Terrence O’Donnell and Toledo Appeals Court Judge Judith Ann Lanzinger — both hold comfortable leads.

[OnTheOtherHand]

22. thanks, although I see why Tfc would like regional splits there, too...

Statewide, both races were polled to be close, and both races were close. But is it startling that Connally did relatively better than Kerry in Republican counties? I wouldn't think it is surprising given that judicial elections are officially nonpartisan and down-ticket (so I would expect Connally's support to be more uniform than Kerry's), but I can't place a probability on it.

If someone wanted to pursue this, the Dispatch might be willing to provide regional splits on the judicial races, for what they are worth. (I would rather work on more direct evidence about the presidential race.)

[Time for change]

23. Thanks for pointing that out minvis

Given that Moyer had a mere two point lead over Connally just a few days before the election, I have to say that it doesn't appear to me that Connally doing better than Kerry in some very conservative counties doesn't sound terribly suspicious to me. For all the talk of her lack of name recognition, she was polling quite well against an incombant judge, and this provides an independent reason to believe that she might have outpolled Kerry in some places on election day.

As I said before, I do believe that Kerry was cheated out of many votes in southwestern Ohio (Not least because of the Warren County "lockdown". But I don't believe that the Connally "anomaly" should be among our top priorities in pursuing this.