As I noted briefly this afternoon, I've found some discrepancies that are in some ways more serious and unexplainable than the ones I reported last night here. I consider this to be a smoking gun -- not for fraud -- but for data that is hopelessly irreconcilable and worthless for ascertaining the outcome of this election.
There are three precincts in Issaquah that together have more people who voted than there are ballots that can be located. In other words, either some voters' ballots disappeared, or people who didn't cast ballots made it onto the list of people who did. Neither is acceptable, but I suppose the former is more likely to have happened than the latter. Precincts 05-2602, 05-3464 and the unincorporated "Gilman" all vote at the Clark Elementary School (no other precinct does). Unlike some of the similar discrepancies I found with polling place ballots, this discrepancy can't be explained by adjusting with other precincts from the same polling location.
1. Ballots vs. Voters at the Clark Elementary Precincts
| Precinct | Absentee Ballots |
Absentee Voters |
Provisional |
Provisional Voters |
Polling Ballots |
Polling Voters |
| GILMAN | 159 | 159 | 5 | 4 | 71 | 70 |
| ISS 05-2602 | 816 | 817 | 62 | 64 | 349 | 345 |
| ISS 05-3464 | 686 | 694 | 47 | 57 | 307 | 300 |
| Total | 1,661 | 1,670 | 114 | 125 | 727 | 715 |
| Precinct | Absentee | Provisional |
Polling |
| GILMAN | 0 | 1 | 1 |
| ISS 05-2602 | -1 | -2 | 4 |
| ISS 05-3464 | -8 | -10 | 7 |
| Total | -9 | -11 | 12 |
| Count | Absentee | "Add-on"/ Provisional |
Polling |
Total |
| Initial Count | 1,661 | 115 | 726 | 2,502 |
| Machine Recount | 1,661 | 114 | 727 | 2,502 |
| Manual Recount |
? | ? | ? | 2,501 |
| Voters | 1,670 | 125 | 715 | 2,510 |
The punchline then is that there are 7 voters from these precincts whose ballots disappeared.
The reason I consider these missing ballots such a huge deal is that the county's explanation for the 3,539 vote discrepancy as an incomplete data entry project doesn't address this problem. Adding more voters to the list will not help find the ballots that disappeared. These 7 inexplicably ballotless voters are not the only examples of voters for whom ballots cannot be located. There are hundreds of precincts whose voter/ballot counts are off. This is merely the largest and most vivid illustration I've come across so far.
We've already seen the counting and twice recounting of nearly 900,000 pieces of paper. In addition to the explained additions of ballots, the number of pieces of paper increased by 336 and again by 59 and without explanation. The reason to reconcile the counting of ballots against the counting of voters by precinct and ballot type is as a check against outcome-changing errors and fraud. If the number of ballots doesn't match the number of voters somewhere you have to assume that somebody made a mistake or that somebody intended to alter the outcome. If you can't reconcile, you have no way of knowing whether you have an honest outcome.
If a group of precincts has more voters than ballots you have to ask how this could have happened. The only explanations I can think of are that : (1) A number of voters were erroneously recorded as having voted when they didn't vote. (2) ballots disappeared and weren't counted. (3) ballots were mixed up in the counting center and placed with other precincts. (4) The ballot counts are bogus.
Any of these problems throw into question the entire work product of the three vote counts. If you are happy to live with the possibility that the outcome of the election is erroneous if not fraudulent, then don't worry, be happy. If you care about the principle that the will of the people be respected, and you want to verify that we had an honest outcome, then we have to look for the source of the 7 vote discrepancy, and make any necessary corrections. The county's approach of reconciling the voter list has to take as a given that the number of ballots is correct and only the number of voters can be wrong. If there are more voters than ballots in a precinct, but the county's assumptions the voters will have to be thrown off the list. By the whole point of the exercise is to satisfy ourselves that the number of ballots was correct. Thus the only way to proceed would either be to go through another count, this time recounting both all of the ballots and all of the voters. Or throw out the mess and start over. Which would you prefer?
Posted by Stefan Sharkansky at January 04, 2005
12:53 AM | Email This
And who could fault them for using this technique for electioneering, because outright fraud in the good-ole-USA is a tough pill to swallow, even for the most partisan conservative.
However, now they have painted themselves into an inescapable corner, because there is no way to make the number of ballots equal to the number of voters on the voter roll without having obviously done so, especially in precincts that are receiving "Sharklight."
And the reason why this is such a smoking gun is that the concept of the number of voters matching the number of ballots in an election is something that everyone from a five year old all the way up to a Supreme Court justice like Barabara Madsen can understand.
ReVote.
Posted by: Jeff B. on January 4, 2005 05:43 AMExactly. And most rational voters are asking themselves, WHY? Rational Washingtonions want a good explanation for WHY this is so in an election with only a 129 vote margin and obvious discrepancies that could total great than 129.
Rational Washingtonions want to know WHY this is hard for some to understand.
Rational Washingtonions know WHY a recontest would provide a clear winner, and are not afraid to let a recontest choose our Governor.
Posted by: Jeff B. on January 4, 2005 06:01 AMI now understand why some of the other County Auditors put forth this "circle the wagons" support letter for Dean Logan. Apparently many are in the same boat. Is the Washington State Assoc of County Auditors a CLUB that cheerleads for members in trouble without first checking on the merit of the charges OR is it a PROFESSIONAL ORGANIZATION? It looks like for some of these County Auditors, it is a CLUB.
1) Congrats to all the County Auditors who cared enough about the integrity of this election to reconcile TO THE PENNY or at least make the effort before certifying.
2) Shame on the Auditors who failed to reconcile.
3) Double-shame on the County Auditors who are trying to make us believe it is ok to be off by so much.
Some of the County Auditors have set the bar on accountability lower than whale droppings. This is pathetic and unacceptable.
Posted by: Mr. Cynical on January 4, 2005 06:02 AMFinancial institutions (for example) use rigorous quality control measures, including balance reconciliations and audits, because if they did not, people would rightly question whether they are allowing fraud to take place -- or participating in fraud. Sensible people expect this to be done in any process where the outcome is critical, especially in any process designed, managed, or carried out by people who might conceivably benefit from subverting the process. This is supposed to happen when your money is banked, when the medicines you take are produced, and when the airplanes you fly on are designed, built, flown, and maintained.
It doesn't take a genius to see that this should happen with the voting process, too. Although my main concerrn is this particular election, the nagging question I am left with is: among the people who made the laws and regulations for Washington elections, which of them benefited from the miserable, Keystone-Cops election process we now have?
Posted by: Boonie on January 4, 2005 06:30 AM In a word you are awesome man.The state gop
owes you big time keep it up.
Hugh Hewitt's If It's Not Close They Can't Cheat
ie; If it's close someone can and will cheat. Looks like Shark has uncovered the Masters of the Game.
Great work Stefan. I have enjoyed reading your posts and commend you for all your effort. Let all Americans hope it will NOT be in vain!
Posted by: Chuck on January 4, 2005 07:38 AMCould the missing ballots have been disqualified (overvotes, etc.)? How do those ballots figure into your count?
Posted by: David Bolotin on January 4, 2005 07:56 AM1) Independent verificiation of the totals
2) Certainty that every vote was counted
3) Certainty that every vote was legal
Many ideas have been floated for 3, so I won't go into that.
For 1, when a voter submits his ballot, he provides one copy to a Republican and one copy to a Democrat. There's a unique ballot number on the ballot, which can be used to verify that the identical ballot was included in both totals.
Both parties tally up the votes separately, compare the results, and if the error is too large, nail down every last discrepancy.
For 2), the voter takes home a paper stub with the same ballot number, and he can look online to see if that ballot number was counted.
It would not be cheap, but it is obvious that the honor system does not work.
Realistically, I don't think anyone can declare who won. The results have been a statistical tie all along. I don't know that Rossi won anymore than anyone can intellectually declare that Gregoire won without insulting the intelligence of the rest of the citizens. The only answer is a revote with a statewide audit that reconciles votes versus voters. If Gregoire doesn't go along, and a court declares the election nullified, she's politically sunk. If a court doesn't nullify, she's sunk anyway, again at least politically.
Even the Leg. Demos are going to be reluctant to circle the wagons with her on anything other than the most benign of issues, knowing how their own political fortunes are inextricably tied to her lack of validity in the public's eye. She may win a battle here but the entire party could lose the war in a very big way if they don't come to their senses and agree to a revote. It's really the only option that can save them. If she gets elected in a reconcilable revote, then she has her mandate. Otherwise, she's a 4 year lame duck with no political capital.
Go with the revote Christine, it's your only logical choice.
Posted by: Bill on January 4, 2005 08:14 AMFoundational to the entire arguement that this site is championing is the fact that Sam Reed could have prevented much of the 'mischief' that may have occurred. The above site is a letter to Sam Reed from Bob Williams which I think is very informative.
"EFF President Bob Williams said the problems in the 2004 election transcend the governor’s race. “The impact of this election on the integrity of the elections system and the governor-elect cannot be overstated. Free and fair elections are the cornerstone of a legitimate representative government,” said Williams. “Party politics should never trump the right of a qualified citizen to cast a ballot and know it will be counted.” "
Posted by: carl on January 4, 2005 08:33 AMThe five year old, perhaps...Madsen, there's some doubt.
Posted by: South County on January 4, 2005 08:37 AMCould you kindly further describe what you call "Absentee Ballots" vs. "Absentee Voters"? Does "Absentee Ballots" represent the number of ballots that were created and sent out to voters? Does "Absentee Voters" describe the number of ballots received back? Or does "Absentee Voters" mean the number of ballots received back and counted? Or does "Absentee Voters" mean the number of valid absentee votes?
A fifth explanation is that additional ballots were sent out that were not properly recorded in the "Absentee Ballots" count. These might have been to voters who did not receive their ballots for example.
Have you been able to ask officials who supplied you with your numbers questions about these totals?
Posted by: Daniel K on January 4, 2005 08:45 AMStefan,
I'm looking at your tables and I'm not sure I follow where you come up with the number 7. I'm not doubting your math, just don't see where the 7 figure is derived. Thanks.
Posted by: Daniel K on January 4, 2005 08:50 AMRobyn
Posted by: Orange Robyn on January 4, 2005 09:20 AMPretty simple explanation.
Then there's the 1 other missing vote. 1 vote? Out of more than 2,500? Talk about making a mountain out of a molehill. There's no fraud, no error, no nothing here. Better luck next time.
Jeez...
Posted by: Nelson on January 4, 2005 09:42 AMIf the same thing is true in King County, it would mean that your tally of "Provisional Voters" would be the voters registered in a precinct who voted provisionally, but in many cases those votes would be cast in other precincts. Similarly, the "Provisional Ballots" tally would contain the ballots of many voters who were registered in other precincts. Comparing the two tallies would be meaningless.
This is just one example of why you can't infer too much from a detailed comparison of the preliminary voter list with precinct reports.
I'm not saying that the possibility of a serious discrepancy doesn't exist. I'm just saying that you can't tell until a detailed reconciliation and updated voter database are available.
And we've got to keep the pressure on for getting a revote - you've done an amazing job looking thru all of this info! And because of you, the local news last night showed a Survey America poll that showed Washingtonians overwhelmingly want a revote and dont want Dino to concede. This is actually a poll I can respect since it was a phone poll.
Posted by: Lauri on January 4, 2005 10:03 AMDead on. Any important process needs autentication and auditing. Humans will take advantage of any system that they no they can.
What's particularly upseting about what Goldy is saying is that he's basically fine with any errors as long as they are justified by laws. This asolute reading of the law is ironically quite contrary to the relativistic reading of law that Democrats usually prefer. I guess the law should be interpreted however it best benefits Democrats.
Goldy is trying to justify through specific enacted law the outcome of a close election, even when there were severe irregularities that may have compromised that election. Much worse, he's ignoring the natural law of fairness that offends the sensibilities of any objective observer that can clearly see that this election result is innaccurate.
Headless Lucy, while it may be true that there is some hypocrisy in Repbulicans wanting the election to be over during earlier counts that were just as flawed as this one, the same hypocrisy is now being exhibited by Democrats now that the result is in their favor as compared to before when Greogire called the election a tie.
Furthermore, Headless Lucy, then we had not yet uncovered all of the evidence that we know now. Republicans are not doing what Democrats are doing. Republicans realize that even Rossi could not honestly take office in this election without either a retally of the election after ferreting out voterless ballots and ballotless voters, or a recontest of the whole election.
We are not afraid of an honest outcome, are you?
Posted by: Jeff B. on January 4, 2005 10:20 AMAny court would disagree with that interpretation.
By analogy, Marc Rich did not commit tax evasion. After all, Clinton pardoned him, right?
Wrong. He's still guilty. He just can't be sent to jail for it.
And the election board does not have the power of pardon anyway.
Posted by: Bostonian on January 4, 2005 10:37 AMRobyn
Posted by: Orange Robyn on January 4, 2005 10:45 AMPat McCarthy of Pierce County elections, reports to Dave Ross on Kiro 710 [January 4] that no one has asked for the list of those who actually voted on election day. She says there is an active voter list and an inactive voter list. Are some of the over 8000 ballotless voters going to be reconciled by these "inactive" voter lists? Any thoughts? Ms. McCarthy also indicated that those who are raising concern over these numbers "just don't really undersand the complicated process" (paraphrased).
I don’t know about Pierce or other counties, but King County’s press release on December 30 makes it clear they were asked for the list of people who voted:
Dec. 30, 2004 Updated at 1:40 p.m.
Preliminary list of registered voters not finalized
Yesterday, in response to several public records requests, King County Elections produced a preliminary list of registered voters who participated in the Nov. 2, 2004 General Election. The data file has been provided to several parties and individuals who had requested it.
The Elections Office notified those receiving the list that it has not been completely reconciled with data from the 2004 General Election.
Is there anyone in the Republican Party who is attempting to find evidence of illegal votes?
Is there anywhere volunteers could go to assist in such an effort?
Have the Republicans actually failed to make plain requests to the other counties, so that they get the list they need in order to make this effort?
Can someone further describe what is being called "Absentee Ballots" vs. "Absentee Voters"?
Does "Absentee Ballots" represent the number of ballots that were created and sent out to voters?
Does "Absentee Voters" describe the number of ballots received back and could this include ballots re-sent to voters that complained they didn't receive their ballot? Or does "Absentee Voters" mean the number of ballots received back and counted? Or does "Absentee Voters" mean the number of valid absentee votes?
Finally, how is the number 7 being derived from the tables in Stefan's post, or is this number coming from somewhere else? Adding and subtracting numbers I come up with other numbers, but not 7.
Thanks.
Posted by: Daniel K on January 4, 2005 10:56 AMI’ll take a stab at answering your questions.
Stefan’s Table 2 shows the net effect of the numbers in Table 1. Add up those totals in Table 2, and you find that there are 8 more voters credited with having voted than there are ballots which were tabulated in the precinct returns.
Stefan then notes that one of the voters whose signatures weren’t in the digital database for comparison was in one of those precincts – thereby offering an apparent explanation for the absence of one voter’s ballot in the tabulation. 8 minus 1 equals 7 voters whose ballots seem to have “disappeared.”
“Absentee ballots” refers to the category in the precinct returns – that is, the ballots which were submitted by the voters, not the number of ballots mailed out prior to the election to the voters.
“Absentee voters” refers to the category of voters in the precinct returns – that is, the voters who are credited with having voted in the election by absentee ballot.
What's particularly upseting about what Goldy is saying is that he's basically fine with any errors as long as they are justified by laws. This asolute reading of the law is ironically quite contrary to the relativistic reading of law that Democrats usually prefer. I guess the law should be interpreted however it best benefits Democrats.Goldy is trying to justify through specific enacted law the outcome of a close election, even when there were severe irregularities that may have compromised that election. Much worse, he's ignoring the natural law of fairness that offends the sensibilities of any objective observer that can clearly see that this election result is innaccurate.
Again, Jeff B, if you have a problem with what I'm posting, why don't come back over to my place and post your comments there, where I'm sure to see them? Unlike some blogs, I never block people from posting.
As to my legalistic analysis, what I am trying to point out is A) these discrepancies are neither unusual nor unexpected, and B) they do not provide legal grounds for setting aside this election. It is my opinion that under current statute, with the current information available, this election should stand.
I have never implied that errors are fine by me, and have strongly urged that we use this controversy as an opportunity to study the election and propose reasonable reforms. I have also never said that these results are "fair." I believe Stefan would agree that I have been consistent in stating that the margin of victory is too far within the margin of error to confidently determine the winner.
But neither statute nor common law require a perfect election, regardless of how close it might be, because nobody can reasonably expect perfection. The courts simply cannot, and should not, set aside an election because the margin is smaller than the suspected number of errors. With no statutory definition of how close is too close, we would set a terrible, terrible precedent where all close elections end up in the courts.
This is not about what's fair, no matter how much that may offend your sensibilities. This is about what's practical. If the election had ended in a tie, the winner would have been determined by lot. Is that fair? Is that democratic?
Now, if there is evidence of fraud or misconduct or illegal votes -- or even gross negligence -- and it can be proven that this favored Gregoire, then that is another story. But I haven't seen such evidence yet.
Posted by: David Goldstein on January 4, 2005 11:20 AMThe law in Washington is (was, until changed from the bench) you need to get at least 1% of the vote to be listed. Ruth Bennett got less than 1% in the primary but then sued and got some judge to declare the law unconstitutional. As the sitting AG, Christine Gregoire could have appealed that decision but didn't.
Oh... and just wanted to point out that Sam Reed chose not to have the AG appeal, because he didn't want to delay the printing of absentee ballots. My understanding is that he was most concerned with military ballots going out on time.
But yes, the Libertarian candidate should not have been on the ballot.
Posted by: David Goldstein on January 4, 2005 11:27 AMIs is still unclear to me how much can be deduced from these numbers. Do we know that these numbers actually reflect what was counted in the certified numbers, and precisely how? How do these number correspond to votes for either candidate? When are these numbers produced (as the ballots are counted? after all the counting?)? Could the totals include mistaken double entries?
Posted by: Daniel K on January 4, 2005 11:33 AMYou cannot declare a winner before you have all these things cleared up. Until these things are cleared up, you do not know what the results are.
***
I hope this is the debacle that ushers in tighter standards for quality control in US elections, not to mention a federal requirement for photo ID at the polls.
What WA has does not serve democracy.
Posted by: Bostonian on January 4, 2005 11:58 AMWhat will hopefully come of this is that the scrutiny will force these descrepencies to be accounted for in full.
We'll see.
Posted by: Daniel K on January 4, 2005 12:09 PMThe process of understanding the discrepancies would turn up votes that should not have been counted at all, and that can affect the results, and likely has done so.
Posted by: Bostonian on January 4, 2005 12:42 PMYou say you have strongly urged for election reform, and to some extent that is true.
But in your partisanship of hoping that you can turn this snapshot moment where Gregoire is ahead within the literal bounds of the law, you are pretty much discredited as someone who objectively wants to reform this obviously broken process.
If you were really honest with yourself, you would be calling for a revote just like we are. For me its gone way beyond partisanship because it is obvious that if the tables were turned, Rossi would be just as suspect as Gregoire. As I have said many times, I am truly fine with a Gregoire win, as long as it is one everyone agrees was accurate.
This is a unique moment in history not because poorly written laws allow Gregoire to become governor even with skepticism in the eyes of the majority of WA voters, but because while the attention for all of this is white hot, we can use this momentum to further the cause of objectivity.
As for not posting on your blog, I do that for two reasons. One, I would just be perceived as a bating troller like so many who post here even if I had the decency to act rational as you have done with your comment.
Two, frankly the level of discourse is greatly elevated and rational here. Do a comparison of your readers comments vs. these here over several posts. The profanity and invective of many of your posters, both left and right is laughable. That's why I disagree with your use of profanity (which I've noticed you have wryly toned way down lately) is that as the host, you set the tone. Not that Stefan has been perfect either. For a better example, see Flags Of The World, Timothy Goddard's blog. You can indeed write whatever you want, and even in profanity, I support your right to do as you want, but you catch way more bees with honey.
The bottom line though is that the majority of Washingtonions (as sceintifically polled) simply are not convinced by your argument. More people realize that there is enough uncertainty with the result that we are far better served by a revote than in just trying to whisk either candidate in with the results as they stand.
I wish you would pay more homage to this important point, even if you maintain your partisanship to Gregoire which I believe you should. This would enhance your credibility and it just might help get us to a result we can all agree upon.
Posted by: Jeff B. on January 4, 2005 12:49 PMI suspect a very small minority of those polled have heard the type of argument Goldy and others have been making, and rather have based their opinion on rhetoric and invective, or less than conclusive and incomplete, information. I'm not denigrating opinions, just wondering what the voters would think if they sat through an exhaustive courtroom like presentation of all the facts, or simply considered that a re-vote without changes made to correct these issues would probably result in the same ol' same ol'.
Posted by: Daniel K on January 4, 2005 01:01 PMI'd bet you at least 75% of the WA voters saw this much happen:
1) The machine count showed a small margin.
2) The loser demanded a hand recount and lost that also, but by a smaller margin.
3) The second recount changed the results, and gave the ex-loser a tiny margin.
4) The ex-loser declared victory, and the election officials, who belong to her political party, certifed the results.
Right there, that is enough to make anyone suspicious. You're fooling yourself if you believe otherwise.
Posted by: Bostonian on January 4, 2005 01:14 PMYou see, you just proved my point. I suspect you are above average in your knowledge of what happened yet you say:
1) The machine count showed a small margin.
2) The loser demanded a hand recount and lost that also, but by a smaller margin.
No, the loser of the initial count did NOT demand a hand recount after the initial recount - or did NOT lose the hand recount that occurred as the second recount, after the machine recount, by a smaller margin. I'm actually not sure what you were trying to say there. You either meant to say machine recount instead of hand recount, or meant to say the hand recount resulted in a change of the outcome.
Until the final outcome, after the hand recount, there was no winner or loser. This was never a best of three counting process. The final recount trumped any previous count. The Rossi camp has been very astute in trying to make their case by constantly stating otherwise.
Posted by: Daniel K on January 4, 2005 01:28 PMAfter the machine count, there was indeed an APPARENT winner, Dino Rossi. Sam Reed designated him Governor-Elect pending the Dems coming up with the money for the hand recount. So, in effect, he was the winner until such time as the hand recount could be afforded and completed.
Now, as the people in the State see it, none of these counts matter as the entire issue has been obfuscated by the further finding of additional ballots in each count, unreconcilable voter versus ballot counts with there being apparently more ballots than voters (bad ratio!), court battles, etc. Regardless of whether or not the public understands all the arguments that Goldy or you and others make, their general opinion has been stated via the survey and an ultimate win by CG will largely be viewed as an unacceptable conclusion. This will hurt her, and the Democratic party immensely.
You may win this battle; time will tell. However, if you do through any means other than a revote or runoff, the repercussions will likely be huge. Win the battle, lose the war. This is a real political pickle the Dems find themselves in. If nothing else, it's fascinating to learn from.
Posted by: Bill on January 4, 2005 03:51 PMUm, because a win that small triggers an automatic recount?
Your nic suits you...
What this spells, and what we should ALL GET BEHIND, is the need for legislation requiring transparent election processes with easily verifiable results (paper ballots, if nothing else).
Unfortunately, the Repub majority is burying just such a bill in committee that would solve the problem.....The "guy who penned the legislation" for voter-verified paper receipts is Congressman Rush Holt of New Jersey. Hilary Clinton sponsored the companion bill in the Senate
The name of the bill is THE VOTER CONFIDENCE AND INCREASED ACCESSIBLITY ACT.
Conservative Washingtonians, of all people, should embrace this move.
Posted by: Lea Jones on January 13, 2005 10:11 AMWhat this spells, and what we should ALL GET BEHIND, is the need for legislation requiring transparent election processes with easily verifiable results (paper ballots, if nothing else).
Unfortunately, the Repub majority is burying just such a bill in committee that would solve the problem.....The "guy who penned the legislation" for voter-verified paper receipts is Congressman Rush Holt of New Jersey. Hilary Clinton sponsored the companion bill in the Senate
The name of the bill is THE VOTER CONFIDENCE AND INCREASED ACCESSIBLITY ACT.
Conservative Washingtonians, of all people, should embrace this move.
Posted by: Lea Jones on January 13, 2005 10:12 AM