Miller v. Lieutenant Governor Craig Campbell et al

Filing 95

MOTION for Leave to File Amicus Curiae and Memorandum of Law by Thomas A. Lamb. (Attachments: # 1 Brief of Thomas A. Lamb, # 2 Exhibit A)(PXS, COURT STAFF)

Miller v. Lieutenant Governor Craig Campbell et al

Doc. 95 Att. 2

Affidavit of William Peck State of Maryland Harford County
)

) ss. )

William Peck, being first duly sworn upon oath, deposes and states as follows under penalty of perjury: 1. I have personal cognizance of the matters set forth herein, and hereby verify that the same are true and correct to the best of my information and belief. 2. I am a volunteer on the Joe Miller campaign, and I have been a write-in ballot observer for several days, usually 2-4 hours at a time, for an estimated total of 8-10 hours. I was also one of two supervisors on the floor who provided the final decision for Joe Miller. I was in the supervisor role about 55% of the time or slightly more from Saturday, November 13 through Tuesday, November 16 (when the write-in counting ended) 3. This affidavit discusses the procedures (or lack thereof) during the counting of the write-in ballots in Juneau, Alaska from Weds, Nov 10 to Tues Nov 16. I believe this information is critical to understanding not only what happened in at the write-in counting location, but it has larger implications as well to the election. The larger implications that were played out during the write-in counts were a) irregUlarities in the election, b) a rush to the finish line, c) protest votes, and d) media malpractice. 4. Irregularities in the write-in counting procedures - summary The standard for the write-in procedures was established in the document "Counting Write-in Votes - US Senate Race" (Attachment A). This was posted on the home page of the Division of Elections' website during the write-in counting operations, but I have not been able to find it online since the counting ended. Some of the issues I describe do not directly affect the vote count, but they do indirectly affect the vote count, and in substantive ways. So in that regard, j do feel the inconsistent application of the write-in procedures directly affected the vote count. The irregularities / inconsistencies in administering the procedures are pretty clear. In general, the procedures were loosely adhered to, and in my opinion, often ignored. Even when I challenged how the procedures were being administered, the responsewas "everything is challengeable", which both Gail Fenumiai and Mike Barnhill (a Department of Law lawyer) told me was the basis for what I considered frivolous challenges by the IVlurkowski team, including one ballot that said "Fuck Lisa" (it probably also had her name as well, so to be clear, the ballot probably also had "Li.sa Murkowski" on it.) I also believe this particular ballot went into the "Challenged Not Counted" pile. I think it is also ironic that my detailed review of the "Counting Write-in Votes" document was done at the suggestion of the lead lawyer on the floor for Lisa Murkowski, Scott Kendall. I had printed out the instructions prior to heading to Juneau, but had only read the first paragraph. So (I believe it was) on Saturday night, I read

Dockets.Justia.com

the write-in guidelines with interest and in detail. And it really made me conscious that the procedures on the floor did not match my understanding of the written guidelines. One thing I think is worth noting is that the write-in procedures document is not very professional, in my opinion, especially for an event of this magnitude. There is no date on the document, no official seal or logo, not even any reference to the state of Alaska. Also, one key point is that it seems clear to me that, based on the inconsistent enforcement of the write-in procedures, I personally conclude that the election workers were not properly trained in their duties. I speak from experience in leading soldiers through five years active service and three years in the national guard, as well as many training exercises in my role in software development I software project management. I do however, want to compliment the election workers for their dedication, honesty. sincerity. and spirit of friendliness and fairness in the execution of their duties.

5. Sorting and counting ballots
The first paragraph of the instructions gives an overview of the procedures to follow:

Counting Write~ln Votes ... VS ~enate Race
When separating ballots to count the individual write~in votes, the ballots will be
initially Borred first. Mtel' the initial sort, if necessary, the director will make n
detcl'.nlination on voter intent. Once ballots are sorted and challenges are
handled, the ballots will be counted and individual write~in results will be
recorded.

Although this point might seem a bit picky, I believe it is worth noting that in the "initial sort", the election workers are supposed to make a determination as to whether the write-in name is a "variation or misspelling of Murkowski (or Lisa Murkowski)". So, in effect, the election workers ALSO make a determination of "voter intent", using the standard of "variation or misspelling of Murkowski".

6. Goldbelt security, lack of security during 5% audit on Weds, Nov 17

Ballots will be delivered by Goldbelt Security. There will be a chain of cUHtody
indicating the district and number of boxeo being trarH~ported from the Division
of Elections ballot l'oom to the Alaska Utho Building and fronl the Alaska Litho
Building to the Division of Elections ballot 1'oom, There win be a division
employee following the ballot transport at all tiules.
Goldbelt Security will remain at the counting center to ensure ballot ~ecudty.
Since I had read the guidelines with interest on Saturday night, this section on Goldbelt security was what struck me on Weds, Nov 17 as I observed the 5% audit taking place. I noticed that there was no Goldbelt security and that ballots were placed in a civilian car and driven off by an elections worker who I presume was alone. This is documented more fully in a separate affidavit.

7. Initial ballot sort - l a c k of consistency Here are the procedures followed by my discussion. There wlU be 5 box lids u~ed to sort ballots, Use the box lids to lnitic>lIy tiort the ballot.<; as follows: 1. Place ballot~ where the oval]S mal'k~d (colored in; X, Star, Chet:k etc,) n~xt to a candidate's name that is printed on the ballot (those othel.' than the Write-In category). 2. Place ballots where the US Senate Race is left blank (no oval marked) 01' l' N(l 0v'4'! more than one oval is marked, or a name is written in, but no oval is ~ W t/ (ivAI,) marked 3. Place ballotB whr.re tl1f' OVAl i~ marked ror WrHe..ln collegOl'y (colo1'ed in) X, Star, Check) and the name iu written <'IS LISA IvIlJRK01VSI<I OR MURKOWSKI and spelled correctly that are NOT chnllenged. 4. 'Place hallot.'i where the oval is 111l:!rked tor Write-In \:tilcgory (colored in, X,

l

Check) and THE NAME \VRITTEN A.-PPF",\ RS TO DE A VARIATION OR MTSfoJ1'ELLTNG OF MlJRKO,,\\'SI<I OR LISA
MUHKOWSKl. Also plnce in this lid any bn.l1ot that <In 0hs~rvt;';r

Star~

dmlkngcg so th<1t {h~ dirl)(Aor (;fli1 mak,.' }] ddemli..llatioD.

5. Place ballots where th~ oval is marked tfJr Write-In category and the name written in is NOT Murkowski) Lisa Murkowski or a variation thCl~cof.

7.a Changing standards in regards to the initial sort When I first started as an observer (on the first day, Weds, Nov 10), we were placed behind the election workers, so we were standing behind them looking over their shoulders (starting at gam). Around 10:45 we were re-positioned to the opposite side of the table as the election workers. This was actually much better, but for a while I was a bit disoriented as I now had to read the write-in ballots upside down. I made a point of how my brain had been locked into the over-the-shoulder view, now I was in the upside-down view. After a few minutes, I got to the point where I could read the ballots upside down with good enough accuracy, so it didn't really bother me after that.
In subsequent days (on different tables), we would sit in chairs opposite the workers, and the ballots were placed in the boxes right-side up for us (hence, upside down for the election workers), but they had already reviewed the ballots. Table 1 (my initial table) sorted the ballots as described, with one exception: they did not consider whether or not the name "appears to be a variation or misspelling of Murkowski". What they did was simply present the ballot for the Miller observer to review / challenge immediately, without considering whether it was spelled correctly (as outlined in the procedures). When I reviewed ballots as the lVIiller observer on other days (at other tables), the ladies would do the initial sort as described by the procedures. That made things a lot easier, as then all I had to do was review the ones they deemed spelled correctly, and every so often I would challenge something they put in the good pile. One point to note is, what exactly would not qualify as "a variation of Murkowski" ? Pretty much ANYTHING went into Box 4 to be challenged, even "Lisa M". While one can surmise the person was THINKING Lisa Murkowski, how can one conclude that "Lisa M" is a variation of "Murkowski" ? Besides that, was there another "Lisa M" on the write-in candidacy list? How about simply "Lisa" ? The point is that different tables implemented different procedures. For someone like myself who visited many tables, it does provide a different experience whether you are reviewing every ballot or only those that have been screened by the election workers.

7.b The initial sort is misrepresented by the Juneau Empire This procedure at Table 1 (showing every ballot to the Miller observer without determining whether it is a "variation of Murkowski" ) is confirmed by a media story that totally misrepresented the review process. The way it worked is that the worker on Table 1 would see that the ballot had a write-in name (that was clearly not for another candidate), then he would present the ballot so the Miller observer could clearly see it, holding it up to eye level so that we could make our decision to accept or challenge. So essentially every write-in ballot was presented in this manner, regardless of whether it was spelled correctly or not, then the Miller observer would make his decision.
The Juneau Empire had a front-page, above the fold picture of a perfectly spelled ballot being presented for review, with the caption "questioned ballot". The clear implication, confirmed by a Juneau Empire reporter, was that the Miller camp was questioning (challenging) a ballot that was obviously spelled correctly. The Juneau Empire reporter confirmed to me that the "message" was unfortunate, that he had spoken to an editor or someone about it, and that it had been corrected for subsequent editions and online. But the damage was done - the correctly completed ballot was on the front page with the clear implication that the Miller camp

was questioning this. The was part of a pattern of misinformation on how the process was working, all to the
detriment to the Miller campaign and correspondingly to the benefit of the Murkowski campaign.
7.c. Election workers determine whether the write-in name is spelled correctly

4. Place hallot.<.;" where the oval is marked tor \X/rite-In ~al<;gory (colored iu, Xl
Stlir, Check) and T.H~ NA1\Jlli\¥Rm'EN APPEARS TO DE A
VARIATION OR MISSPELLING OF IVIURKOWSKl OR LL'}A
I MnH.KOWSKl. Also place in 1:his l1d f.\ny ballot that ~n 0hs~rvt:r
dl::l1knl{c~ ~o tlwt ~h>;: dlr(:(..tor e8:i1 maki~ (l defeEl11natio[\.

When 1read this Saturday night, the thing that crossed my mind was that the election worker was ALSO
in the position of determining "voter intent". But more specifically, the standard is actually different than
voter intent, it is "a variation or misspelling of Murkowski". While this generally is not a difficult exercise, it
actually becomes quite tricky when considering the wide range of spellings that we all saw. Is "Lisa M" a
variation? how about "Makoskie" ? how about "Muskisk", which I witnessed as a GOOD vote for Murkowski
. (as announced by Gail Fenumiai). So the point is that the practical rule was that anything that was NOT clearly someone else and that had a last name starting with "M" was placed in Box 4 (the challenge pile, from which Gail would pull from to make her pronouncements.) 7.d. Sorting ballots into Box 2 (no oval or double votes)

2. Place ballots where the US Senate Race is left blank (no oval rum-ked) or X.
nlOre than one oval is marked, or a nmne is written inl but no oval is

rnl'll'ked
-~~--------------- L:./:::,-?~----~_.

4. Place balIoL.~ where the oval is m~l'kcd tor Write-In cakgory (colored in, X
Star} Check) and THE NAt"-IE\VRITTEN A P P R A f(S TO HE A

VAHIATION OR MISSPELLING OF IVIURKOWSKI OR LISA

MUHKOWSKI. Also place in this lid any ballot that ~m d1alkng~~~ ~o tlwt ~h(: dit'(j<;·tOl' ~J2Hl mak(,: (} deferil1 illat1ol) .

obs~rvt;";r

During my in-depth review of the procedures, I focused on the ballots that should go to Box 2 ('no oval' votes and overvotes). My understanding is that these ballots go to Box 2 and are never heard from again, since they are clearly not valid. Box 4 is only for the ballots where the oval IS filled in but are a "variation of misspelling of Murkowski". This is
the box from which Gail pulls from to do her review of the challenged ballots.

Director Detetmination ~-""""'--".'''-'')
1, Each ballot in b~x lid #4 will be reviewed to deterrnin oter inte~y When
reviewing these ballots, they will be sorted into 4 categor e5aS1OUows:

So the Director is to pull from Box 4 (and not Box 2), hence my observation that "no oval" votes
and "overvotes" should never be heard from again (with one exception). The one exception is that the

election worker is to "place in this lid any ballot that an observer challenges so that the director can make a determination" (see the last line of point 4 in the graphic just above (with the instructions for Box 2 and 4). So, to follow a trail of an overvote, it is supposed to go in Box 2 (and never heard from again, as I understand it). However, they (along with all "no oval" votes) would end upin Box 4. Then Gail would accept as valid votes for Murkowski many overvotes (as I have previously documented). For example, a ballot with the oval filled in for Joe Miller but with an X over the oval, along with the write-in name, was considered a vote for Lisa (I personally witnessed this. Gail's response was (paraphrased) "you could tell by the other ovals that a properly filled out oval is what the voter wanted, so since Joe's has an X over it, that voter doesn't want Joe." (But note that an X through the oval is considered a correctly completed oval.) We know from Gail's statement and from a clearer understanding of the numbers that at least 1,541 of the 2,016 CNC ballots were "no ovals", which as I've indicated my understanding is that they should have gone into Box 2 (and never processed beyond that). Another example of an overvote going for Lisa is the one where the voter filled in the oval for Joe in black, then in blue ink scratched out his name and wrote "changed my mind" along with a vote for Lisa (in blue ink). This was accepted as a valid vote for Lisa. This is documented in a separate affidavit (with picture). As a direct contradiction to the vote for Joe with the X over the oval, there was a vote for Scott McAdams but also a vote for Lisa, but with an X over Lisa's oval, yet this was considered as a vote for Lisa (documented separately with a picture). So, not only were overvotes handled inconsistenly, they were in general handled improperly, as I understand the guidelines. Although "everything is challengeable" is a principle that the Murkowski team followed and the Division of Elections endorsed, this is not the kind of thing that breeds confidence in the process, at least from my perspective. Finally, here is another key point on the "Box 2" rules:

Aftel' initial sort raise hand for director, 01" director designee. The director's
designees are division staff appointed by the director to validate that the ball(
in box lid #2 are truly blank or over voted ballots. All other ballot..'1 will be
~'eviewed by the director.

l

As I understand this, the director (or designee) only checks that Box 2 contains the "no oval" or "overvotes". "All other ballots will be reviewed" implies to me that in fact, Box 2 votes are NOT pulled for review by the Director, unless for whatever reason the Murkowski team challenges them. I have documented in a separate affidavit my opinion on why the Murkowski team challenged ballots that clearly did not /lleet the standard for an acceptable ballot (the main reason being that we theorized they wanted the "Challenged Not Counted" number to be substantial, so that it appeared as if the DOE was reasonably rejecting ballots, when in fact the actual percentage of challenged ballots that the Director did !'JOT accept for Murkowski was miniscule).

7.e Director determination of voter intent, by reviewing ballots in Box 4

Director Determination /":#,.-.-'_.......)"
1. Each ballot in box lid #4 will be reviewed to detennin~~ h.:tte!!t. 'When reviewing these ballots, they will be sorted into 4 categories as fullows:
This point indicates to me that ballots in Box 2 are not to be reviewed as part of the Director's review process,
unless a ballot designated for Box 2 is challenged. Although my recollection is that Box 2 and Box 4 seemed
interchangeable, I do recall that Gail would simply come to a table and pull from Box 4 to begin her review. But
in virtually all cases, overvotes and undervotes were part of the directors review when she pulled from Box 4.
"No oval" votes were always rejected by Gail, but then challenged by the Murkowski team, and they went
in the CNC pile. "Overvotes" were usually counted for Murkowski, regardless of the circumstances. I have
documented separately three overvotes that were called for Murkowski, and this was pretty much the norm.
On Sunday, November 15, I made it a point to question either "no oval" votes or "overvotes" being placed in
Box 4 (instead of Box 2) For example, as ballots were being sorted by the election workers, their instructions
were to put "no ova'" votes and "overvotes" ballots in Box 2. I doubt that all Murkowski observers (over 15
tables) challenged these ballots consistently during the sorting process (the result which would have been for
them to be placed in Box 4). That is why my sense / belief is that it was pretty fluid between Box 2 and 4, and
that essentially both boxes were the source of the ballots that Gail reviewed.
While on Table 5 pretty early in the morning, I made it a point that I wanted to clarify what was supposed to
go in Box 2. I even asked that the proceedings be halted in order to clarify. That request was dismissed out of
hand, and we ultimately moved on after a brief but kind of tense situation.
On the very next table we moved to (Table 1), I again questioned the sorting of the ballots, again trying to
make the point that both "no oval" votes and "overvotes" were to go into Box 2 and never be heard from again.
I was not even allowed 30 seconds to make my objection, this denial coming from Gail and Mike Barnhill.
This is documented more fully in a separate affidavit. Plus this emphasized the principle employed by the
Murkowski team and endorsed by the Division of Elections that "everything is challengable", and in fact,
overvotes often ended up as "challenged counted" for Murkowski.

7.f Counting the rejected ballots ostensibly for Lisa Murkowski

4, For the ballots in box lid #5 (write-in v o t e for candidate other than
Murkowskl) sort the ballots into the separate names. Count and recotd
the individual results for those candid~tesappearing on the certified
write~in candidate Jist or on the ballot (record the results next to the nalne
on. the fsults sheet). If the name written in docs not appear on the results
sheet record the votes on the Hne Other Write-In.

Ballots for write-in candidates other than Murkowski were to be placed in Box 5, then recorded separately in their vote totals. For example, Joe Miller got 20 write-in votes (which, by the way, is the highest number of write-in votes for anyone other than Murkowski). Only 620 out of 103,805 votes were rejected as being not for any candidate. That means that at most, only 0.6% of the write-in votes were deemed not to have been for Lisa Murkowski (or some random candidate, like Bart Simpson that I say). It also means that in one form or another, 99.32% of votes not clearly for another candidate or were either accepted as a vote for Murkowski by the Division of Elections or challenged to be a vote (but not counted by the DOE) by the Murkowski team. This is a much higher acceptance rate than previous write-in elections in Alaska, as Nick Begich described shortly before November 2. (Her acceptance rate was actually 97.38%, 89.52% uncontested and 7.86% accepted by the DOE but challenged by the Miller campaign. 7.g Joe Miller's write-in votes. Joe Miller has 20 write-in votes. But it is unlikely that these votes are included in his current total of 90,740. This is because the page where his vote total of 90.740 is reported is also the page where the total number of write-in votes is shown as 102,252. But we know that the true number of write-in votes is actually 103,805, which is reported in a different location of the system (and this is the issue behind the "1,541 vote discrepancy", i.e., two parts of the system that should be reporting the same results but do not). 90,740 link: http://www.elections.ataska.gov/results/1 OGENR/data/results.pdf 103,805 link: http://www.elections.alaska.gov/results/1 OGENR/data/resultsWI.pdf Joe's totals may actually be 90,731, as reported to me by a reliable source who (I believe) added up the numbers by precinct (or by district), coming up with an entirely different total. The point is there is no reliable spot to go and find the answer. While Lisa will still be in the lead by 2,000 votes or so, it is nevertheless frustrating to look at three different reports and see three different results. 8. Changing standards in regards to taking pictures and media coverage It wasn't until around Saturday that our team was allowed to have cameras. By then a lot of the media damage had been done, as we could not counter hearsay with photographs. The famous "Murcowshit" ballot (accepted by Gail Fenumiai as "voter intent" for Lisa), slipped by without a photograph, because we were not allowed to take photographs on the first day (or for several days thereafter). To be fair, the name wasn't clearly spelled that way, but it was clearly "cow", clearly an "h" instead of a "k", and after the "i", the writing swooped up higher than the "i" with the cross of the "t" being evident. This is documented more fully in my "misspellings" affidavit.

Partly as a result of not being able to take pictures, a Fox News reporter either saw firsthand or was told that our team was "challenging perfectly good ballots", and he reminded me of such. My response was (paraphrased), "there's probably a good chance that a perfectly good ballot is in the challenged pile, but this is not only rare, but it is not our policy to do so." I explained to him how each morning we would meet and emphasiz.e what the standard was, and that we should adhere to the standard. Because of our consistent adherence to the standard as outlined in the law, the Anchorage Daily News reported that a Murkowski team member said that (paraphrased): "The Miller team is not challenging frivolous misspellings." But the point is that the lack of taking pictures resulted in this unsubstantiated accusation by the media that we were challenging perfectly good ballots. Then this message is passed on up the chain to Fox, and that's what Geraldo Rivera uses as his opening challenge to Joe Miller to question what was going on. It was either Sunday or Monday that I noticed John Tracy on the floor taking pictures with his cell phone. We had been told that pictures could only be taken by 1 of 2 floor supervisors, or by the observers themselves. I was confused as to why Mr. Tracy was on the floor taking pictures, in my mind, he didn't exactly fit the mold of campaign staff or volunteer. I inquired as to whether he was a floor supervisor, pointing directly at him to make sure it was clear who I was asking about (reporters were also allowed on the floor). I believe it was Gail who confirmed that John was, in fact, one of the two Murkowski observers.

9. Rushing the process to get finished - on the floor
The conduct on the floor by Gail Fenumiai demonstrated that the larger effort was to rush the election to conclusion. This was demonstrated by her continual rushing from table to table to begin the review of the challenged ballots. Whenever a table was finished processing a precinct, they would raise their hand, and one of Gail's assistants would note down the table #. Most of the time this was Mike Barnhill, a lawyer from the Department of Law. When Gail finished a table, she would quickly get the next table number from Mike, or whoever was tracking the queue at the time, and she would hustle over to the table, frequently without respect to whether the Miller final reviewer was present. I noted to several colleagues that if I paused for even a moment to reflect or chat with a teammate, she would be off hustling to the next table to begin the count, and then I would have to myself hustle over to be there for the review. It wasn't until the very end, sometime Sunday or Monday, that they made an effort to see if I or Matt Johnson was coming to do the review. This was simple professional respect but to me it was quite evidently absent the first several days.

9.a Rushing the process to get finished - s t a r t i n g on November 10 instead of November 18
The original announced date to begin counting the ballots was November 18, then this was changed to November 10. I'm not sure when either the original or revised dates were announced, but i do remember thinking on November 4 that we had 6 days to assemble and train a team in Juneau. Counting ballots in Juneau poses severe logistical challenges to both candidates. Had this been done in

Anchorage, many of the challenges we faced would have been eliminated. One simple example is the deleterious effect on the overall process by conducting this in a building that was not actively being used, without internet access, and far from the base of both team's operations in Anchorage. We were able to keep the 15 tables fully staffed over the seven day period of counting ballots, but it was a struggle due to the limited support base in Juneau compared to Anchorage. Team members had to be vetted, trained, and cared for. We operated on a shoe-string due to not having internet access at the building, no lunch facility, and with most of the core team being from Anchorage and living out of a hotel. In the seven days of counting, we had Sunday morning off, then also Tuesday morning. Even getting laundry done was a challenge. We racked up personal expenses during this time, including high costs for rental cars (though this should be reimbursed). None of these issues would have risen had the count been done in Anchorage. We would have been able to easily fully staff the tables, without disrupting people's lives as was done with the deployment to Juneau. We would not have incurred travel expenses (airfare to Juneau), local transportation costs (rental cars), nor lodging costs. Our team could have been more easily trained and cared for in Anchorage, and our ability to ensure proper command and communications would have been significantly enhanced. On Monday evening, it was announced that the final day of counting would commence at 11 :30, and we would "work until we were done", with the unstated but clear message that this would be 5pm or even later. Yet, around 3:30 in the afternoon (and inclUding a lunch break), we were finished with all the write-in ballots. At the conclusion of the counting on Tuesday, this is when we were blindsided by the 1,541 "extra" votes that gave Lisa an unexpected (from our perspective) and substantial lead of around 2,200 votes, which I have documented separately. The explanation for the 1,541 was presented matter of factly. As noted separately, the explanation that the 1,541 votes were due to "no oval" votes (that are Challenged not Counted [CNC] ) while at the same time Lisa's vote totals went up unexpectedly by virtually the same amount. The explanation didn't match the facts as we knew them. Lieutenant Governor Craig Campbell publicly stated (on the Mark Levin radio show) that the purpose of moving the dates forward by 10 days was so that "the people could learn the winner [as quickly as possible]." As it stands now, it will be well past November 25 that this election is resolved, yet it is quite possible that the preparation time prior to the original date of November 18 might have resulted in a more eqUitable and efficient outcome. 9.b Counting military I overseas ballots on Weds, Nov 17, not Friday, Nov 19 as reported by Gail The final day for receiving military / overseas ballots was Nov 17. I was told personally by Gail that these ballots would be counted in Juneau on Friday, November 19. Since we knew that it looked like Lisa would have the lead (barely) in uncontested votes, we held out hope that we could gain ground with the military / overseas votes. Wednesday was the last day for counting the write-in votes, so we pondered what might happen with the military / overseas votes. We were also told it was "about 1,000 votes". It turns out that the votes were counted on November 17, but (I believe) they were not posted until November 19, the final day that vote totals were increased. It turns out that there were 718 military / overseas votes

(between the top 3 candidates), of which Joe got 292 (41 %), write-in's were 236 (33%), and Scott McAdams received 190 votes (26%). Lisa ended up with 214 out of the 236 write in votes (an acceptance rate of 90.68%, 1.16% higher than the overall write-in rate of 89.52%). So we did not make the gains in the military / overseas ballots we had expected. And so Weds, Nov 17, was the final day of counting all votes, and the day we got blindsided by the 1,541 votes, giving Lisa around a 2,200 vote lead. I noticed that Kevin Sweeny was immediately giving a recorded interview to some reporter. This was also the day that Lisa Murkowski returned from Washington, giving a press conference to announce her supposed victory. The next day, November 18, was the celebration of Ted Stevens, on his birthday, held at the Discovery Theatre at the Alaska Center for the Performing Arts. 10. Protest votes The first sign we had of a protest vote was the infamous "Murcowshit" example. This happened on Day 1 (Weds, Nov 10), before cameras were allowed by the candidate's teams. Again, the name wasn't written in clear letters, but it was definitely "Murcowshit". Later, other really butchered variations of what might appear to be "Murkowski" made me realize, or believe, that what I was seeing wasa significant number of protest votes. Again, the "Fuck Lisa" vote seems to be a pretty clear indication of a protest against Lisa (yet the Murkowski team challenged this, and it ultimately ended up in the "Challenged not counted" pile, as I recall). I also saw a vote for Murkowski with a sad face, something like this: So what is the voter intent here? Gail called it for Lisa, but I challenged this, so it is in the CC number, but it is still officially a counted vote for Lisa, as the Division of Elections considers it. So it simply became evident to me that many, many votes were actually protest votes. Despite the protestations of many, it is really not that hard to spell Murkowski if one wants their vote to count. The more and more I looked at really awful "variations" of Murkowski, the more I was convinced that it was intentional in many cases. But of course I cannot prove that, no more than Gail prove what was the voter's intent. So it was kind of comical to me what she considered a vote for Murkowski. Often, Gail would huddle with her lawyers to determine what direction to take. For me, it took about 2 seconds to determine if the ballot met the standard. The only issue was bad hand-writing, which in more cases than not, I gave the benefit of the doubt to the voter. It was only when it was clearly not in compliance with the law that I challenged it. This standard led to an air of fairness that was patently evident throughout the floor, from the Murkowski floor supervisor to the election workers, and to a certain extent, among the press who remained to the end.

Q

In regards to protest votes, there was certainly plenty of opportunity to spell her name correctly, including through the use of the write-in lists present at the polling place, in addition to wrist-bands, pens, cards, and ads depicting a spelling bee. And I became more and more convinced "that noforily is'voteYinterifthe wrong standard to use, it is simply folly. Especially when "Muskisk" is deemed to have been "intended" for Murkowski. Finally, I want to point out again my "granny exception rule" (documented in another affidavit), which is not meant to be light-hearted but to emphasize that the Miller campaign is about doing the right thing. Further affiant sayeth naught. Dated this Zr4-lt day of November, 2010 in Abingdon, Maryland

William Peck

Subscribed and sworn to before me, a notary public in and for the State of Maryland this :7 November, 2010.

17 0~ of day

'.......
: I

..
.\

....

.'~

""",

...

',0- - ,,) , ·· : , \\.

..

Counting Write-In Votes - U S Senate Race When separating ballots to count the individual write-in votes, the ballots will be initially sorted first. After the initial sort, if necessary, the director will make a determination on voter intent. Once ballots are sorted and challenges are handled, the baD.ots will be counted and individual'write-in results will be recorded. Ballots will be delivered by Goldbelt Security. There will be a chain of custody indicating the district and number of boxes being transported from the Division of Elections ballot room to the Alaska Litho Building and from the Alaska Litho Bnilding to the Division. of Elections ballot room. There w i11 be a division employee following the ballot transport at an times.

Goldbelt Sectitity will remain at the counting center to ensure ballot security.
One house district vvill be counted at each table. The voted ballots are :in separate, sealed envelopes by precinct. There will be only one precinct opened at a time. When a precinct count is completed, fhe voted ballots axe returned to the nrecinct envelooe and sealed bv the counting: team. Challene.:ed ballots will be sealed tTl separate envelopes, identified with the district and precinct nu.mber.
.1.
.L.

J

V

U

Results will be reported by district at the completion of each district count,
r ,. o Be1 \ Y ·oS illl Jttilln or
"1

~ne

process:

I11itial Ballot Sort Th~re vlrHI be 5 box licL used to sort baUots. Use the boxIids to ;1l1tially 80rt Ute ballots as fallows: 1. Place ballots \-\there the oval is marked (colored i.n.; Xi St2.t', Check etc.) next to a cand.idate's name that is printed on the bullot (those other than the vVrii:e~In category). 2. Place ballots where the US Senate Race iB 1eft blank (no oval marked) or more than one oval is marked, or a name is written in, but no oval is
IT!.EUked

3. Place baHots where the ovn! is rnal'ked for V.frlte-Irt Ci:lteg-orv (colore.d i:r:, )C, , . Sr-a.r. Check) and the na..rne i:; written as LISA T"'fU KO'\"ISKIOR
~ ~

:MURKOWSKI and spelled correctly that ar .

.L

OT challenged.
".

4. Place ballots wrere the OVEll is nwrked IO!' \A,;'ritc-In calegory (colored in.." X.. ....... '" Star, Check) and THE N.!u.~IE 'NRITTEN APPEARS TO BE A VARIATION OR IvUSSPE-LLING OFl\HJltKO\VSKl OR LISA
lY UnAV)·Y'::'X\J..
~,~y"rt","&...·r~T''t"r~..,r·r:

/".I~lU

,~'f_.

fHCll:e 10 W\S iHl any (1r.I.(;1 in,,'.
:-"i.

1! __

~

: ··

"';4 '_

t~.:

_.

1._1't .. _ .. 1. . . .

~111. i10:':;~·l·Ver

_

. i.

. ...

chn1jr:nsC~ :~D -i.·h~li t1'H~ d~T~{.~·tf'~r ~~X!. ~~~;:1t~f7.

~,t~1·~?"'C('!.~j~~:~i·:!~~ ..

5. Place ballots where the oval is marked for Write-In category and the name written in is NOT Murkowski, Lisa Murkowski or a variation thereof. After initial sort, raise hand for director, or director designee. The director's designees are division staff appointed by the director to validate that the ballots in b o x lid #2 are truly blank or over voted ballots. All other ballots will be reviewed by the director.
Diredor Determination .~ 1. Each ballot in b o x lid #4 will be reviewed to det~..JWhen reviewing these ballots, they will be sorted into 4 categories as follows: a. Determination made as vote for Write-In candidate Usa Murkowski and determination is NOT challenged. (These ballots will be placed in b o x lid #3 for counting.) b. Determination made to count the vote for Write-In candidate Lisa Murkowski and determination is challenged. These ballots will be segregated and placed into envelope labeled - C h a l l e n g e d Counted for Murkowsld. The total number of votes from these ballots will be recorded on the results sheet as votes for Murkowski - C o u n t e d Challenged. c. Determination made to NOT count the vote for Write-In candidate Lisa Murkowski and determination is challenged. 'These ballots will be segregated and placed into envelope labeled - C h a l l e n g e d NOT Counted for Murkowsld. The total number of votes from these ballots will be added to the results sheet as Murkowski - N o t Counted Challenged. d. Determination made to NOT count the vote for Write-In candidate Lisa Murkowski and determination is not challenged. These ballots iJJj;/1I ~..t../ are then placed in box lid #5. 6YV-jj;/ r
j

;-, ,t· tL/ v

Recording Results
1. Hand-eount the number of ballots from box lid #3 (votes for Murkowski) and record the total number on the results sheet (certified. write-in candidate list) for Lisa Murkowski. 2. Count the number of ballots in t h e C h a l l e n g e d - C o u n t e d for MurkowskilT envelope and record the total number on the results sheet line Lisa Murkowski - C o u n t e d Challenged. After counting, return these ballots to the envelope. 3. Count the number of ballots in the IIChallenged - N o t Counted for Murkowski" envelope and record the total number on the results sheet line for Murkowski - N o t Counted Challenged. After COWlting, return these ballots to the envelope.
II

Page2of3

4. For the ballots in box lid #5 (write-in vote for candidate other than Murkowski) sort the ballots into the separate names. Count and record the individual results for those candidates appearing on the certified write-in candidate list or on the ballot (record the resultS next to the name on the results sheet). If t h e name written in does not appear on the results sheet record the votes on the line Other Write-In.

Page 3 of3