Dean Logan misstated the law about "certification" of election returns
When Dean Logan was questioned today by King County Councilman Hammond about the meaning of “certification” of the election returns, Logan said that it would be a violation of state law to refuse to certify the returns by the fifteenth day after the general election.
Since Logan is a recognized expert on election laws, perhaps someone will ask him what RCW 29A.60.200 means when it says, “if they can be ascertained with reasonable certainty” –
What Hammond was driving at was the difference between an election in 1999, when a man who had been in Logan’s position refused to certify election returns that couldn’t be reasonably ascertained – and when a clear written statement of the discrepancies was provided to the canvassing board.
In 2004, there were many hundreds of ballots that were improperly included in the vote counting process in King County, yet the canvassing board certified that the numbers they reported as their election returns constituted a “full, true and correct representation of the votes cast.”
Clearly, the members of the canvassing board couldn’t have ascertained the returns “with reasonable certainty.” Mr. Triplett accompanied Logan today, and stated that the total numbers of ballots exceeded the number of voters by approximately 1200 at the polling places and by approximately 400 in the absentee voting. That’s too big a discrepancy for them to ignore.
They might not have known of the magnitude of the discrepancy between the number of legal votes and the number of votes counted. It seems clear that Logan didn’t even disclose to the other members of the canvassing board the full extent of the discrepancy between the number of ballots included in the counting of votes and the number of qualified electors who cast ballots in the election. When asked by Hammond whether Logan’s office prepared any report resembling the one done for that 1999 election, Logan said unequivocally that they had not.
Since Logan is a recognized expert on election laws, perhaps someone will ask him what RCW 29A.60.200 means when it says, “if they can be ascertained with reasonable certainty” –
The county canvassing board shall proceed to verify the results from the precincts and the absentee ballots. The board shall execute a certificate of the results of the primary or election signed by all members of the board or their designees. Failure to certify the returns, if they can be ascertained with reasonable certainty, is a crime under RCW 29A.84.720.
What Hammond was driving at was the difference between an election in 1999, when a man who had been in Logan’s position refused to certify election returns that couldn’t be reasonably ascertained – and when a clear written statement of the discrepancies was provided to the canvassing board.
In 2004, there were many hundreds of ballots that were improperly included in the vote counting process in King County, yet the canvassing board certified that the numbers they reported as their election returns constituted a “full, true and correct representation of the votes cast.”
Clearly, the members of the canvassing board couldn’t have ascertained the returns “with reasonable certainty.” Mr. Triplett accompanied Logan today, and stated that the total numbers of ballots exceeded the number of voters by approximately 1200 at the polling places and by approximately 400 in the absentee voting. That’s too big a discrepancy for them to ignore.
They might not have known of the magnitude of the discrepancy between the number of legal votes and the number of votes counted. It seems clear that Logan didn’t even disclose to the other members of the canvassing board the full extent of the discrepancy between the number of ballots included in the counting of votes and the number of qualified electors who cast ballots in the election. When asked by Hammond whether Logan’s office prepared any report resembling the one done for that 1999 election, Logan said unequivocally that they had not.
posted by Micajah | 1:43 PM
12 Comments:
”Mr. Triplett accompanied Logan today, and stated that the total numbers of ballots exceeded the number of voters by approximately 1200 at the polling places and by approximately 400 in the absentee voting."
I didn't hear that Triplett statment.
How do you reconcile that statement about "1200", with the 216 excess ballots shown in the reconciliation spreadsheed and summary?
That's a huge difference. Shouldn't the reconciliation report control with respect to certification. And if it does, then there was an approximately 99.89% accuracy rate, which seems easily sufficient to justify certification of the county's result.
The statement by Triplett was made during his opening remarks. It caught my attention, because it's the first time I've heard any King County official state that part of the discrepancy between the number of voters and number of ballots resulted from absentee ballots that were included in the vote count.
Recognizing that Logan stated today that there is "some overlap" in the numbers, note that these are the numbers available:
216 more ballots than voters are completely unexplained in the reconciliation report;
348 "PVAB" -- prov'ls reported by poll workers to have been run through AccuVote machines at polling places; and
660 "Adjusted #" to remove from the reported "variance" the number of prov'ls thought to have been run through the AccuVote machines because there is "no label" envelope to be found for that number of prov'ls issued.
Those numbers total 1224 ballots which were inserted into the vote count under "irregular" circumstances at the polling places which left King County without a way to verify who cast them.
As for the accuracy of the reconciliation summary, note that at least 8 precincts that used the St. Benedict polling place were omitted from the summary. (See comment posted by Greg at Sound Politics at 5:34 PM, March 13.)
Since the reconciliation summary didn't total the columns for the "Election Day MC" and "Add-ons" it's not apparent that precincts (and thus ballots) were omitted from the reconcilation summary.
So, it's too early to know just how much weight to give to the information released by King County on Friday afternoon. It will take a while to go over it and see what else they have omitted.
When I said this, I should have said "without a way to verify that they were cast by qualified electors before they were included in the vote count" --
Those numbers total 1224 ballots which were inserted into the vote count under "irregular" circumstances at the polling places which left King County without a way to verify who cast them.
Some of the 348 (252 is the number I've heard reported) have been identified as having been cast by qualified electors.
Micajah,
I still don't see how you can reconcile that with this statement from the reconciliation summary.
"Following the reconciliation process, a degree of variance remains. Polling places that showed a negative variance have a sum total variance of “-158”, indicating that the sum of provisional ballots issued, add-ons, and the AccuVote machine tabulation was 158 less than the sum total of ballots signed for in these polling places.
Likewise, polling places that showed a positive variance have a sum total variance of 216, indicating that the sum of provisional ballots issued, add-ons, and the AccuVote machine tabulation was 216 more than the sum total of ballots signed for in these polling places.
The combined negative and positive variance is 374.
Based on 334,185 polling place ballots cast, this represents a variance of .0011, or just over one-tenth of one percent."
From this statement I conclude that there were only a total of 216 more votes recorded than signatures in all the poll books. In addition there were 158 more signatures than votes recorded in those same poll books.
This is a far cry from the 3000+ discrepancy that Sharkansky was claiming.
(granted he was including 400 alleged absentees also)
And the net difference of 216-158 = 58 votes, isn't very far off the mythical 20 vote discrepancy that Councilman Irons claims he recalls from the 2000 election, which is held up as the standard of rectitude.
Now how you fit in the 348 provisional votes passed thru the vote machines and this 660 "adjustment" number, I'm less sure about.
chew2,
You are stuck on the 216. The 216 ballots are ballots which appeared out of nowhere. No one has any explanation where they came from. They could have been forged copies of legitimate ballots for all anyone knows. Move away from the 216.
When the King County report talks about the "variance," it doesn't mean what ordinary people mean. It means only that there is a variance between the ballots known to have issued and the ballots in the vote count. It doesn't use "variance" to refer to the variance between the number of qualified electors known to have voted and the number of ballots in the vote count.
In other words, King County is still trying to get everyone to take their eyes off the fact that far more ballots got into the vote count than there were qualified electors who were known to have cast those ballots.
The 660 "adjusted #" is nothing more than ballots which were not legally cast at the polls -- and which the canvassers deduced were issued as provisional ballots.
Since the "reconciliation" being done by King County was only an effort to figure out where the pieces of paper called ballots came from, they subtracted 660 when they thought they had figured out where the people who inserted those ballots into the AccuVote machines got them.
The primary reason there is such a difference between the "variance" being reported now by King County and the variance between voters and ballots noted by Sharkansky is that King County is still not talking about the variance between voters and ballots. King County is now talking about the variance between ballots known to have been issued (whether as regular or provisional ballots doesn't matter -- just that they were issued) and ballots that were included in the vote count.
Those 660 are ballots which are not supposed to be in the vote count, because they weren't legally cast.
216 plus 660 equals 876 ballots in the vote count which weren't legally cast.
Logan said today that there is "some overlap" of the 348 illegitimate ballots and the 660. I don't know much of an overlap there is. Until Logan said that, I had thought all 348 were included in the 660. (I haven't gone over the reconciliation report in detail, so I don't know if the numbers on it add up to show that all 348 are in the 660 or not.)
chew2,
I happened to look at the last page of the reconciliation summary document and noticed that the 348 provisional ballots which reportedly were inserted into the AccuVote machines at the polling places are completely separate from the 660 "Adjusted #" ballots which were deduced to be provisional ballots inserted into the AccuVote machines.
In other words, there were 660 + 348 provisional ballots that were improperly inserted into the AccuVote machines. When Logan said today that there is "some overlap" in those numbers, he was incorrect -- unless his own written reconciliation summary is incorrect.
That leaves the 1224 number I noted before as the minimum number of illegitimate ballots that were inserted into the AccuVote machines at the polling places. That's 216 ballots that came from nowhere, 660 provisional ballots deduced by the canvassers as having been inserted into the machines, and 348 provisional ballots that were reported by the precinct workers as having been inserted into the machines.
That's the minimum, since the reconciliation summary omitted all the precincts at one polling place that I know of -- and there may be other omissions. If there are other omissions, then there may well be more provisional ballots or ballots from nowhere that went into the machines.
Micajah,
Here's my math. There were 216 more ballots than signatures. There were also 158 more signatures than ballots. So a net of 58 more votes than signatures.
Provisional voters signed the poll books before being issued a ballot. So of those who signed, some 660 probably deposited their provisional votes directly into the voting machines without being checked. I'm still not sure what those 660 constituted. If they truly are all "no label" ballots, then you should be able to do a check by adding up all the "no labels" notations in the reconciliation notes pdf file and they should add up to around 660. (But see below, where some are also pvab's)
I'm not sure whether all the 348 pvab are included in the 660 or not, but in many cases they overlap. Take a look at the Beacon Hill Elementary polling place, 13 pvab and 13 adjusted #'s. Almost all the pvab's from my brief scan of other precincts are matched by equivalent adjusted #'s.
Of those 348 pvab ballots, 83% were said by Logan to have been cast by qualified voters. Whether you can project that high a rate of validity for the "no labels" is harder to say.
So we're left with maybe:
60 invalid pvab's,
Say 400 additional no label provisionals, with say 50% invalid or 200,
58 net ballots without signatures.
That's my preliminary take.
So how bad is that? The one thing that disturbs me is that they weren't able to prevent some 660+ voters from depositing their provisional ballots, even though they had taken steps to see this didn't occur.
Then from the legal point of view, is this enough to sustain an election contest. I don't think so. I don't see misconduct or neglect of duty here. They took reasonable steps to prevent the stuffing of provisionals, but weren't successful for some reason. Moreover, it's unlikey the guestimated 350+ invalid votes here changed the margin of victory.
Granted this is all pretty iffy and preliminary.
ps. You say: "The 216 ballots are ballots which appeared out of nowhere."
For me the most plausible assumption is that these ballots were given out without the voter signing in the poll book.
Micajah,
Oops that reference to "Beacon Hill Elementary" should have been "Beacon Towers" with 13 pvab and 13 adjusted #'s.
Here's the text of the notes:
"Hand counted ballots. Probably 7 pb with no label were put through Accuvote. Probably 4 pb with no label went through Accuvote.
ID - AV jammed once; some pbs could have gone through; had 2 lawyers and 7 pollwatchers there all day (signed SS)"
LOL! 2 lawyers and 7 poll watchers, and still 13 pvba were fed thru.
But some of the other notes don't make sense to me, since other "no labels" are noted also.
chew2,
I don't yet know how to figure out the reconciliation report numbers.
For example, after looking again, I cannot tell whether the 348 are separate from the 660. The columns are totaled separately, but it seems impossible to figure out how many of the 348 provided a basis for the 660 "Adjusted #."
I looked at Aki Kurose Middle School to see if I could understand what they were doing just by having the notes and the reconciliation summary available to me. I can't. More than a dozen "no label" provisional ballots disappeared. Yet their disappearance isn't reflected in the "plus/minus" column. Several "PBAV" prov'l ballots were inserted into the machines, but they didn't get reflected as corresponding "plus/minus" numbers. Without the poll book and ballot accountability form, it seems there's no way to know what the canvasser and the person doing the summary were thinking about that polling place.
I went back and looked again at the Bothell Library polling place, since Sharkansky made the poll book cover sheet available for one of the precincts. It still makes no sense.
Assuming that "Signature 180hc" means that the canvasser hand-counted the signatures in the book, did that number include the 8 signatures which should have been in the book for the 8 prov'l ballots that were issued?
The "WANDA 171" tells me that scanning the bar codes showed 171 signatures were entered in the poll book by people who were issued regular ballots. (Those signing for prov'l ballots wouldn't have a bar code next to their signatures, so nothing could possibly be scanned by the bar code reader.)
The difference between the scanned bar codes and the number of ballots counted was 8. There were 177 ballots counted by the machine and 2 "add-ons" that weren't counted by the machine for a total of 179.
The canvasser noted that 4 of the 8 prov'l ballots were "no label," which ought to mean there were no envelopes (with a stick-on label containing the voter's identifying info) containing those 4 ballots. Those 4 ballots disappeared.
So, the canvasser had 4 prov'l ballots in hand, and had 179 ballots in the machine. That totals 183.
There were 171 regular signatures and 8 prov'l ballot signatures for a total of 179 signatures. If the "hc" was of all the signatures, then the canvasser probably felt the 180 was the correct total of signatures.
With all that, how did the canvasser come up with the idea that there were "+7" ballots in that precinct?
There were 177 + 2 + 4 = 183.
There were 180 signatures.
So, there were 3 more ballots than signatures -- unless the "180hc" referred only to signatures for regular ballots, and not for prov'l ballots.
If there were 180 regular ballot signatures, then the canvasser had fewer ballots than signatures. The 180 signatures were matched by only 179 ballots in the machine. When you add the 8 signatures for prov'l ballots, there were 179 + 4 ballots total, but 180 + 8 signatures. So, you're 5 ballots short.
The canvasser had 4 prov'l ballots out of the 8 issued that were missing ("no label").
So if the canvasser was trying to figure out whether all the ballots issued in that precinct could be accounted for, the "+7" fits nothing.
There were certainly 4 prov'l ballots missing. There was either 1 regular ballot missing or 3 too many ballots or 5 too few ballots.
The only way I can think of to get "+7" is to say there were 180 total signatures (including the 8 who signed for prov'l ballots), and there were therefore 172 regular ballot signatures (not the "WANDA 171").
Then, when the canvasser did the arithmetic the result is 179 ballots in the machine minus 172 regular ballot signatures equals 7 too many ballots in the machine.
Once the canvasser got that "+7", then he or she should have looked to see if there were enough missing prov'l ballots ("no label") to explain the presence of those 7 extra ballots in the machine. (Both prov'l and regular ballots had the same identifying precinct number, so the 179 ballots the machine counted as having come from that precinct came from that precinct.)
But, there were only 4 "no label" prov'l ballots -- so only 4 of the 7 extra ballots in the machine from that precinct could be accounted for as having been issued as prov'l ballots but improperly run through the machine.
Where did the other 3 ballots in the machine come from? They had that precinct's identifying number on them, but there was apparently no record that they were ever issued.
Nothing in the notes or the reconciliation summary leads me to an answer, so even though this might be where "+7" came from, it doesn't fit the facts once I try to use the 4 "no label" ballots to explain the presence of +7 ballots in the machine.
That particular Bothell precinct actually seems to have had 3 ballots that couldn't be explained at all, rather than "+7" which could be "adjusted #" away (but for 1 ballot among the 32 the canvasser said were present for the whole polling place -- I don't have any idea why the canvasser thought only 31 of 32 extra ballots could be explained as "no label" ballots that were run through the machine).
It could be that there is no way to understand the canvasser's work without having the cover sheet notes and ballot accountability forms available.
Or, it could be that the canvassers did a FUBAR job of it, and the reason these few precincts I've looked at don't make sense is because the canvasser's work makes no sense.
This comment has been removed by a blog administrator.
This comment has been removed by a blog administrator.
I deleted two duplicates of the last comment I posted. The software or server was not telling me anything more than "server not found" or "page cannot be displayed" when I clicked to post the comment. Then, the comment didn't show up as having been posted. Apparently sometime during the night the three times I posted the comment finally were posted -- long after I logged off and quit trying. I hope this doesn't become a regular problem!
Post a Comment
<< Home