January 06, 2005
A signed confession

It should have occurred to me late last night when I first read it, but Dean Logan's "explanation" of the thousands of discrepancies between the voter list and the precinct canvass is as good as a signed confession of massive incompetence, or worse.

When it was reported that the voter file released by King County on Dec. 28 showed 3,539 fewer voters than there were counted ballots, King County Election's response was:

King County Election Director Dean Logan said a full, updated list is expected to be complete by the end of next week
Mrs. Fraudoire's reliably, helpfully unbrilliant spokesman Morton Brilliant said:
Gregoire's spokesman Morton Brilliant said Republicans were acting like children watching a batch of cookies baking, reaching in too early, then complaining that they're still gooey.
[Thanks, for the own goal, Mort! The "baking cookies" analogy helps plant the perception that King County is cooking the books!]

So the implication is that the data is merely preliminary and the full, updated list is expected to be complete by the end of next week.

Now that the "end of next week" has turned into the end of this week, and its increasingly obvious that King County Election's Enron-style accounting numbers aren't going to add up, Dean Logan is telling a story that he wasn't telling a week ago:

Historically, the number of votes cast and the number of voters credited with voting does not typically balance.


The process of crediting voters for voting is not designed to determine if voter fraud occurred, but rather a process to ensure voter registration lists are updated and current... .

That's funny. If the numbers were never expected to add up, why has this never been stated until yesterday? Probably because the notion that the numbers aren't expected to add up is completely bogus. Here is what RCW 29A.08.125 says:
The computer file [of voter records] must include, but not be limited to, each voter's last name, first name, middle initial, date of birth, residence address, gender, date of registration, applicable taxing district and precinct codes, and the last date on which the individual voted. The county auditor shall subsequently record each consecutive date upon which the individual has voted and retain at least the last five such consecutive dates.
Obviously, if elections officials follow the law, every voter's last date of voting will be recorded correctly and if every ballot and every voter is processed correctly, then the number of voters will reconcile with the number of voters. It really is that simple.

The claim that election data has historically not reconciled is interesting, if disturbing. But is it even true? More precisely, how closely have election data reconciled in the past? And how different is this year's data from previous elections? One person who posted an anonymous comment on an earlier post claims this that other recent elections have actually reconciled very closely. I can't get behind that claim without being able to document it. If anybody has specific information about how well previous elections have reconciled, please bring them to my attention.

In the meantime, there were already enough unbelievable changes to Dean Logan's original story about the voter list, that he may as well have signed a confession that the results of this election are worthless and should be thrown out.

Posted by Stefan Sharkansky at January 06, 2005 04:08 PM | Email This
1. A simple question - why? Why wouldn't the number of voters credited not balance with the number of votes cast? He just say that historically they don't match, uh, why not?

Posted by: Marc on January 6, 2005 04:24 PM
2. I can provide those documents for you, Stefan. They're simply copies of some of the pages of certification documents that used to be handed out at canvass board meetings back when Republicans Chris Vance and David Irons represented the KC Council on the canvass board from 1997 - 2002. The Dems took over the Council, and thus, the Council canvass seat in fall 2002.

You can probably get them from the election office with a PDA request (provided they haven't shredded them to hide the evidence, or "misfiled" them along with some of their absentee ballots....)

Alternately, just send me an e-mail and we can work out the particulars. And keep up the good work!

Posted by: Annie on January 6, 2005 04:42 PM
3. You can go to the following link and send a note to your state reps. I am sure there are other places to get this, but it is fun changing they belly-aching canned letter to one asking for a revote to get this resolved. Enjoy!


Posted by: Jon on January 6, 2005 04:44 PM
4. If the election was run properly (which is a very big IF in this case), the numbers should balance EXACTLY. KC says we (us peons) don't know how complicated this is... well, tell me, how complicated can it be? Yeah, there are almost 900,000 ballots, but so what. Those are spread over close to 3000 precincts, right? I don't care if there were 5 million voters, each and every precinct should be able to balance their voters and number of ballots exactly. If they can't they should be able to state why, such as 400 Fraudoire "voters" stuffed provisional ballots into the machine before we could stop them.

Dean Logan and company must go, as well as the head man himself... you know, the guy who is responsible for all this chaos, Mr. "tax to the max" Ron Simms.


Posted by: TomP on January 6, 2005 04:50 PM
5. Serious Question:

If the challenge works, shouldn't the entire election be thrown out (not just the governor's election)? Shouldn't we re-vote on each and every issue from November 2nd? Isn't that necessary for consistency and to properly throw out the results from an election which is seen to be illegitimate?

Posted by: jim on January 6, 2005 04:57 PM
6. I don't think it would be necessary to toss the entire election since most of the races were well outside the margin of error...unless that felon tally gets to be really high!

Posted by: Jon on January 6, 2005 05:03 PM
7. Jim,

The answer to your serious question is that any ballot item in which the margin of victory is smaller than the number of illegal ballots (or smaller than the size of the unreconciled difference in voter/vote counts) should be thrown out. In races where the margin of victory is larger, it can be presumed that the errors in balloting did not change the outcome.

Posted by: Moscow Mark on January 6, 2005 05:04 PM
8. Jim, good point. But that might be opening another can of worms, that no one is prepared to deal with at this time.

Posted by: JG on January 6, 2005 05:05 PM
9. Still waiting for a reasonable explanation of why, historically, the voters credited does not match the number of votes. This is simply astounding to me. To admit, basically, our controls are so bad that you can't expect us to actually balance the number of votes to voters do you.

Posted by: Marc on January 6, 2005 05:10 PM
10. Stefan,
Just spoke to the person in charge of the Kittitas Co. Auditor's Office who informed me that reconciling the number of voters with the number of votes counted does not precede the certifying of the results to the SOS. It takes place later as part of a cleanup operation. An effort is mandated, but there's some leeway on the time frame.

P.S. We have now have an interim Auditor. Dave Bowen who some credit with "leading the charge" to support Dean Logan is gone. Apparently his e-mail to encourage the Auditors Assoc. in backing Dean Logan was a parting shot in his leave-taking from office.

Posted by: RLG on January 6, 2005 05:13 PM
11. "Annie" with the intriguing documents from King County -- you neglected to give a valid email address.

Please email me at theshark .At. usefulwork .dot. com

Posted by: Stefan Sharkansky on January 6, 2005 05:13 PM
12. Marc:

You actually expect competance? They are government workers after all (no offense to good gov't workers...of which my father was one RIP). I didn't learn to be accountable for my work until I left government service and started working in the private sector. It's a lot harder to slack off when there is serious accountability...after all, who are they accountable to? Us? Looks like they're shakin' in their boots.

Posted by: megs on January 6, 2005 05:13 PM
13. If you have a spare moment, you might ask your King County Council Critter if Logan's statement

The process of crediting voters for voting is not designed to determine if voter fraud occurred, but rather a process to ensure voter registration lists are updated and current

represents the intent of the Council's policy.

A call or e-mail couldn't hurt.

Posted by: Boonie on January 6, 2005 05:19 PM
14. I would think that even if the result were outside the margin of "error" or "fraud" -- it's still a fraudulent election.

It would otherwise suggest fraud is ok, as long as there isn't enough. Bad message and bad result.

Seems like it's all or nothing....plus, I bet there are plenty of elections within the margin of error isn't the margin ususally 4% or something) throughtout the Nov. 2nd election (local, state, and federal elections)...

Plus, does this affect the primaries? There were some close primary races as I recall...

Posted by: jim on January 6, 2005 05:19 PM
15. I am confused. If there is a revote before the voting rules can be clarified or changed what is in place to insure a legitimate election.

One of the men (can't remember his name) running for DNC Chairman was on Fox news. When questioned about the Ohio debaucle today he said it was important for the dems to hold up certification to make a point that voting regulations need to be changed to make it easier for everyone to vote. If Washington State is an example, how much easier can it be. It doesn't matter who you are, alive or dead or how many times you vote. I think they would like to implement this system nation wide. Which brings me to the point, if Christine becomes our governor she will go along with the dem agenda and we will not see election reform in the State of Washington. IMHO

Posted by: Diane on January 6, 2005 05:28 PM
16. Diane, I'm not sure the rules need to be changed for a revote, because I don't think a revote would be close.

I don't know if changing the rules would be legal, but I am sure they're thinking about that.

Following the rules that exist already would probably reduce the margin of error quite a lot, and I would guess a lot of voters are annoyed enough to switch to Dino. (Some the other way too.)

What do you think, though?

Posted by: Bostonian on January 6, 2005 05:47 PM
17. RLG---Kittitas
I think the reconciliation issue may be debatable to some. The statutes don't say it in precisely those terms. But it sure strongly implies it.
How on earth would you know if all votes were counted and accounted for? It should be done by precindt. Ballots should physically be sorted by precindt to just in case things don't reconcile rather than run ballots thru in mixed mode.

I believe you will see more "circling the wagons" on this reconciliation issue. I know they did it in Jefferson County prior to certification FOR ALL 3 COUNTS. If they can do it...why can't everyone else. I mean how do you know if there was ballotbox stuffing and/or unstuffing going on??
Interesting Sam Reed wouldn't put this on his priority list of changes. Maybe it's because it already is required and he won't admit it?

Posted by: Mr. Cynical on January 6, 2005 06:15 PM
18. Electoral Reforms That Are Not Reforms

Has it occurred to anyone (besides me - I have a nasty suspicious streak, frequently tinged with a bit of paranoia) that the one action that (I believe it was) Huennekens pointed to as evidence of his commitment to cleaning up the process may actually be a diversion?

I refer to his statement that he was shocked - shocked! - to find out this past May that election workers had been routinely counting absentee ballots that arrived that had no signature, contrary to state law. He said that he immediately ordered a stop to this practice.

Is it possible that his real reason for wanting this stopped is that this is an easily verifiable violation, one which could (and ultimately did) focus a lot of unwanted attention on what was going on?

If you are going to run an operation that will reliably deliver the vote to the party in power, the smart way to do it is to use methods that are not obvious, such as allowing people to register with no address (Precinct 1823), or mixing provisional ballots in with poll ballots.

Posted by: ewaggin on January 6, 2005 06:19 PM
19. Changed Overvote Policy

As a follow up to that last thought, what is the latest word on the story (reported here) by one of the people who participated in the manual recount in KC, that a third of the way through the policy on overvotes was changed, so that all such were passed up to the canvassing board for a determination of voter intent (BTW, being Greek or Assyrian or whatever, wouldn't haruspex be spelled "haruspex" in Ukrainian?), rather than being disqualified on the spot by the 3-man teams?

Posted by: ewaggin on January 6, 2005 06:27 PM
20. Mr. Cynical,
I agree that reconciliation should be required prior to certfication. I also accept TomP's idea that this is achievable when it's undertaken at the precinct level, even if there are millions of votes with thousands of precincts: simple accounting suffices. It's clear that some of the rules need tightening.

Posted by: RLG on January 6, 2005 06:40 PM
21. "...as good as a signed confession of massive incompetence,"

That's what he'd like people to think. His focus nonw is to let the media lable him incompetent rather than a designer of election fraud.

Posted by: Baynative on January 6, 2005 07:12 PM
22. Report on King county Elections – May 2004


Provisional ballots. Page 47
-“There is currently no way to identify a provisional ballot that has been counted in error at the polling place.”

-“Some provisional ballots are counted at poll sites in error during each election.”

“Some provisional ballots are incorrectly processed during the canvass period.”

“With the increased turned expected in the 2004 presidential election, there will be an increase in the number of provisional ballots cast.”

“While the correct process requires that provisional ballots should be validated during canvassing before being tabulated and that they should not be counted at polling places, running a ballot through the poll site tabulator is an easy mistake for a voter to make. Poll workers are not always able to watch provisional ballots closely enough to prevent them from being counted.

Security. Page 48
“ Reviews of Elections Section by the Secretary of State and the King County Council’s consultant have highlighted significant shortcomings with basic election security.

“Serious issues have been raised by national computer experts and local activists concerned with the security of our elections, including the activities of the County’s major elections vendor.”

“Security practices have been improved in response to concerns. However, security should be addressed proactively, fixing issues after they arise harms public confidence. Indeed, some observers have expressed concerns about MBOS security.”

Process Control (Computer System). Page 52
“Although unlikely, if a malicious programmer did gain access to the tabulation programming, it would not be difficult to evade detection by the current testing process, because absentee counting is spread out over many days.”

Posted by: Newman on January 6, 2005 07:24 PM
23. If there is ambiguity in the law and also differing practices between counties,we've been wrong in assuming that reconciliation of the numbers of voters and counted votes occurs uniformly prior to certification. This should be the case! As noted, how else can we trust there's validity to the election?

Whatever happens with the re-vote effort, contesting this election is necessary to set in motion the steps leading to real reform. Until the rules are tightened and enforced, an "acceptable level" of sloppiness invites corruption while covering up even worse behavior that works to manipulate the results.

Posted by: RLG on January 6, 2005 07:36 PM
24. "The process of crediting voters for voting is not designed to determine if voter fraud occurred, but rather a process to ensure voter registration lists are updated and current... ."

This statement by Dean Logan bothers me......

If this *process* (of ensuring voter registration lists are *updated* and *current*) has been the true purpose behind attempting to reconcile the votes after elections over the years.....Why are we seeing so many voters listed twice - when they moved prior to the previous election? Why are we seeing people who died before the previous election listed now?

Why does it appear that the current list contains so many voters that should have been purged during prior election *reconciliations*?

I mean heck! In what year does King County intend to update their voter list information? It is apparent - they have not done it for several elections!

This makes Dean Logan's entire statement ridiculous!

Posted by: Deborah on January 6, 2005 07:38 PM
25. Deborah: "This statement by Dean Logan bothers me...."

Me, too. He seems to be saying that the voter lists are maintained for the purpose of ensuring that the voter lists are up to date and correct. And here I thought the voter lists were created to ensure that only qualified citizens voted, and that they did so only once per election...silly me.

Every time I read his explanation, the term "circular reasoning" comes to mind.

Posted by: Patrick on January 6, 2005 08:03 PM
26. Patrick: "He seems to be saying that the voter lists are maintained for the purpose of ensuring that the voter lists are up to date and correct."

Good point. My former pastor used to say "if you can't say Amen, say Ouch!" One thing I haven't been able to figure out is why the Demos keep talking so much. Don't they know "you have the right to remain silent..."

Posted by: JG on January 6, 2005 08:28 PM
27. If the challenge works, shouldn't the entire election be thrown out (not just the governor's election)? Shouldn't we re-vote on each and every issue from November 2nd? Isn't that necessary for consistency and to properly throw out the results from an election which is seen to be illegitimate?

Imagine two scenarios:

(1) In an election, it is clearly provable that two votes have been fraudulently cast (but there's no way of knowing for whom); the final tally is 5001 to 4999. Should the election be thrown out?

(2) In an election, it is clearly provable that two votes have been fraudulently cast; the final tally is 9997 to 3. Should the election be thrown out?

True, both elections are somewhat marred by fraud, but ruling that the second should be thrown out would allow anyone who didn't like the way an election was going to trash the results. The fraud in the second case shouldn't be ignored--whoever perpetrated it should be prosecuted--but it should not affect the election results.

Posted by: supercat on January 6, 2005 08:35 PM
28. Supercat,

What if some of the other race's outcomes fell within the fraud/error margin?

What if it was determined that 8500 ballots were cast in error or by fraud AND they were discovered in an area where there was another close outcome?

Yanking those ballots COULD change the outcome of other close races within the state.....

Why shouldn't that be a consideration?...(Other than the obvious nightmare consequences..)

Posted by: Deborah on January 6, 2005 09:46 PM
29. Would you accept a banking statement that was this screwed up?
I wonder how much of this confusion is a direct consequence of motor voter laws and of the laws that prevent challenging suspicious voters.

Posted by: Walter E. Wallis on January 7, 2005 04:47 AM
30. Well, we are up to 29 comments and I nobody has taken a stab at my question. Why would the number of voters credited, historically, not be on balance with the number of votes counted? This seems like a really simple question. Still waiting for a legitimate answer. Call me crazy, but I think asking for the number of voters and number of votes to be the same isn't asking too much, is it? I can see being off by 10, 20 votes maybe. But 3,000+. Are you kidding me.

Posted by: Marc on January 7, 2005 09:17 AM
31. Marc, I think that the answer is: "They should balance."
Unless of course the system is not set up so that the two could be reconciled, or that it would not serve the powers that be to find out if they balanced, or they were too lazy to balance them, they didn't know how to balance them, or they thought "Hey, What's a few thousand votes or voters more or less, among friends?" Or it was never balanced in the past, why should they start now?

I really don't know how, or if, the Demos in charge of King County think. I encourage you to ask Ron Sims, for the definitive answer.

Posted by: JG on January 7, 2005 09:47 AM
32. Hey, Marc;

From an R in another KC, i.e., Klickitat County,who was an observer in the hand-recount:

We were all relieved that our auditor and staff had proper policies and procedures in place, which resulted in a 0/0 outcome, i.e., no difference between the machine-recount and the hand-recount.

There's a common-sense reason for that. I spoke with one of our county commissioners yesterday, whose job it is to supervise the canvassing board, and was told that the FIRST thing they do is determine whether the votes cast are the same number as those signing up to vote. Duuuh!

They were the same in this election, so the board proceeded with its subsequent canvassing duties. Had they not been the same, the board would have attempted reconciliation before proceeding, during which process they would have determined mistake or fraud.

Our rural KC is much smaller than the urban KC, but that should not matter a whit. Accounting principles don't change at the bank with the size of the deposit!

I suspect King County's electoral process has been fradulent for years and now, perhaps, we will learn the truth. I'm hoping for a court case and evidentiary proof of what many of us have suspected for a long time.

Does this remind anyone of New York's Tammany Hall or Chicago's Daley Machine or LBJ's ballot stuffing, Democrats all?

Posted by: JimmyLane on January 7, 2005 12:52 PM
33. One partial reason I read for the imbalance in KC is that in KC provisional ballots are counted in the precinct where cast, rather than in the precinct where the voter is registered. If this is true, I would be curious to know what means are used to prevent multiple voting via provisional ballots.

If provisional ballots are forwarded to the precinct where a voter is registered, then whoever is processing them can check the voter rolls to ensure that the person has not already voted. If two provisional ballots exist for the same person--even if they come from different precincts--this can be detected. If a privisional ballot is cast for a person who also voted in person, this too can be detected.

Perhaps there is some way to validate provisional ballots outside the "right" precinct, but I don't see how.

Posted by: supercat on January 7, 2005 10:50 PM
34. supercat, No.

provisionals are supposed to be counted in the precinct where the voter is registered.

Posted by: Stefan Sharkansky on January 7, 2005 11:01 PM
Post a comment

Email Address:



Remember info?