* *** POSTMORTEM_2016-11-08 TXT - 13 Nov 2016 18:27:23 - JKNAUTH We had a very good turnout in precinct 19-09 for the 11/8/16 general election -- about 21% at the polling place and 57% via absentee voting (mail-in or one-stop) for a total of 78%. The weather was good on election day and for Monday setup. This was the second election (the first big election) at our new polling place, New Bethel Baptist Church. The polling place again was excellent with great support from the people at the church. Statistics ========== Registered voters: 4493 (as of 11/1/16) M100 vs. ATV counts 934 vs. 934 Provisional ballots: 27 Spoiled ballots: 8 Emergency bin ballots: 0 Challenged ballots: 0 "White Paper" absentees: 2534 Late mail-in absentees: 16 ---- Total absentees: 2550 Write-in ballots: 34 Curbside voters: 15 AutoMARK voters: 0 Observers: 0 Polling booths: 13 (including provisional and accessible) Total chairs at tables: 16 (at three large tables) Transfers out: 5 approximately Transfers in: 3 approximately At wrong polling place: 21 approximately Ballot Count Reconciliation web page: http://jgkhome.name/WakeBOE/Ballot_Count_Reconciliation_2016-11-08.htm Staffing ======== We had only 11 officials vs. the plan of 12, so we often had people doing double duty. Most of our officials were very experienced. Five had worked at early voting this election. Everyone worked very well on Monday and Tuesday. Handling the Large Number of Voters ======== === ===== ====== == ====== There were many voters in a long line at the door when the polls opened. The inside hall leading to our voting enclosure winds around and most of the line was not visible from the enclosure. At one point some voters unexpectedly started moving the back of the line down another part of the hall in an undesirable direction. When we learned of that, we had to get the line rearranged and under control. The initial big burst lasted about 90 minutes. Then thruout the day there was a much smaller but fairly steady flow. The end of the day was quieter than normal for a large election. Most of the 19-09 voters had used early voting. Because of the length of the ballot (40 races), it took a typical voter many minutes to complete. That made booth availability a bottleneck. We were supplied only 13 booths, which were quickly used up. We then set up three large tables with a total of 16 chairs for any voter to use if they didn't want to wait for a booth. This helped a great deal. All the 29 booths and chairs were usually occupied during the early morning period. We had a greeter at the door of the voting enclosure to limit the number of people entering that room. We wanted to avoid having a large bunch of people standing around holding a ballot and waiting for a booth or chair. Pollbooks ========= 1) Unfortunately the pollbooks still had the letter splitting problem that election officials have often complained about. However because so many voters had voted via early voting or absentee mail-in, we did not have as many voters in the pollbooks as usual. We had only four pollbooks for this election, but each was large. 2) The pollbooks did end up pretty ragged by the end of the day. ATVs tended to accidentally rip off after the pages had been turned many times. We had to tape back the loose ATVs. Supplies ======== 1) There was no white Absentee List in pollbook box 1. Everything else on the Packing List was there. After multiple calls to/from various BOE people, they printed a copy (on yellow paper) and delivered it to us before I left for the night. Thanks to the OC for doing all that! This was a huge list (42 pages) and was definitely not something we overlooked when unpacking the pollbook box (we had only one box); that was my original fear. We had spent a lot of time searching for the missing list before calling the Help Line. 2) Neither of our two curbside shields had an instruction sheet in the back sleeve. We used my curbside flow sheet as a replacement: http://jgkhome.name/WakeBOE/curbside-2016-November.htm 3) The new vinyl signs seemed to balloon in the wind (including the wind created by passing cars) and many fell over. 4) We ran short of name tags. Our coordinator brought us more. We put the leftovers in the CJ binder. 5) We received only two "ID Printed on ATV List" sheets. We needed three -- one for Registration Table, one for Help Table, and one for Curbside. 6) The Saturday pickup labels on the ballot boxes did not stick well. I found one box had popped open when I unloaded my trunk Saturday. I often saw in past elections that such labels became unstuck. 7) The site-specific "No Curbside in this Lot", etc. signs worked well to steer people to the right parking lot. We had fifteen curbside voters. See http://jgkhome.name/WakeBOE/BOE_Supplies.txt for my unresolved supplies comments from past elections. Database Peculiarities ======== ============= Our Help Table officials found some likely problems in the BOE database(s). See the associated provisional envelopes for more details. 1) A voter said she lived on Deacons Ridge St. The laptop (and Help line) said Deacon Ridge St. (no "s" at the end of Deacon). The voter said the subdivision was called Deacons Ridge. She also said there were two street signs on that street; one said Deacon and one said Deacons. Presumably there is some official name. The street signs need to be made consistent and the subdivision inhabitants, BOE, etc. must get in sync with the official name, whatever it is. 2) At least two voters had received a recently-issued Voter ID card with a faulty residence address. The card listed one address, but the laptop and Help line listed a different, now obsolete, address. One would think that the Voter ID cards would have been printed from the same database used to create the laptop and Help line displays. It's surprising that the cards seemed to be more up-to-date than the database. 3) Several members of a family (same last name) lived in the same Wake County house. The voter said Internet lookups displayed some of these as Wake County voters and others as Franklin County voters. The "Franklin County" voters were not visible on the laptop or via the Help line and were considered to be non-residents and unregistered. VAST ==== Our Help Table people liked VAST very much, but did not like the loss of the full wildcard facility. Thanks for getting the spaced VRNs and highlighted error warnings implemented in time for this election. Hardware ======== 1) The near-silent error beeps of the M100 caused one of our officials to miss overvote errors on two occasions. Each voter had left before the error was noted so the official had to press Accept to allow the next voter to enter a ballot. 2) The curbside doorbell worked well with the signal easily reaching far inside the building and the bell easily heard over the noise of the voting enclosure (vs. an M100 alarm). Some voters (new voters?) seemed to have problems recognizing the curbside stand and driving to where they could use the button. The A-frame sign by the stand did not seem to help those voters. I don't recommend any changes for the stand or sign. 3) We emptied the steel ballot box after just short of 400 voters had entered ballots; there was a comparative lull at that time. We found the ballots were not anywhere close to causing a stacking problem. They had all fallen to the bottom, although some were pretty crumpled. Given that experience, we did not try to empty the bin again as quickly. Near the end of the day, the rate of incoming voters was very low. We were then only a little over 500 voters beyond the first emptying, so we decided to wait for poll closing. When the bin was opened (it then had about 600 ballots), again there was nothing close to a stacking problem. It looked much like earlier in the day. We did joggle the steel ballot box several times during the day; that might have helped settle the ballots. 4) Having only one laptop caused backups at the Help Table. We often had multiple people waiting to be served. Documentation ============= I had previously provided comments on the current documentation. See the 9/5/16 and 9/7/16 files in the "2016, November Election" section of http://jgkhome.name/WakeBOE/WakeBOE_TOC.htm. On 10/19/16 I was given answers to some of these, but some of those answers did not really address the questions I had asked. On 10/20/16 I sent a reply on this, but got no further feedback. Now that the BOE will have time before the next election, all the previous comments need to be looked into more carefully. Here are some additional comments: Chief Judge Pocket Guide (8/23/16) ----- ----- ------ ----- 1) On four pages of the PG, this edition (vs. the June edition) says the Chief Judge Pocket Guide must be packed in the "Blue Reconciliation Bag". The previous edition correctly said it should be packed in the CJ binder. Why was this changed? As I had pointed out in my 9/5/16 comments, the PG is needed to pack the CJ's car; it has a checklist for that. I assume "Blue Reconciliation Bag" means the blue plastic bag in which the Reconciliation Form is sealed, so putting the CJ PG in that would make it unavailable to use as a checklist for packing the CJ's car. In my 9/5/16 comment I just pointed out the bullet misordering on the page about packing the CJ's car. I had missed that you have the "Blue Reconciliation Bag" statement on the four "Chief Judge Notes" pages. 2) Tuesday Morning, Tabulator section: Add as a final bullet, "Call your coordinator after the tabulator has been set up successfully." Our coordinator has specifically requested this action. I don't know if it is wanted by all coordinators. 3) Tuesday Morning, Interior section, bullet 3: Needs to be reworded (see my 9/5/16 comments). Note it has the implication that there is already an Absentee List at the HT to which the Monday Night Absentee List (yet another name for this) must be "added". The big Absentee List that used to be put at the HT was deleted several elections ago because it was not needed there -- it duplicated what is on the laptop, e.g., "ALREADY VOTED" information. 4) During Election Day, Observe at Help Table section: Each official are -> Each official is 5) During Election Day, Things you can do... section: Delete the "Reconciliation form" bullet since the precinct number has been preprinted on that form for several elections now. 6) Closing the Polls, At 7:30 pm section: Curbside cars might form another line that must be considered. There might need to be two lists gathered and two sets of officials gathering names and preventing more people from getting into the two lines after 7:30. As has been noted before, the RT list must include people at the HT. 7) Packing Up, Unvoted Ballots section: "Place unvoted ballots in their original cardboard boxes." We are given only one empty box to put unvoted ballots in. For more boxes, we use those that get emptied as the BT pulls packs from them. Since we have to unload the steel ballot box every 400-500 ballots and seal the box we put those in, we can run into a situation at the end of the day where unused ballots cannot go back into the exact box they came from originally. We have to seal them in whatever unvoted ballot box is available. This can particularly happen when a precinct has many ballot styles. I assume the "original" here is meant to be interpreted loosely. 8) As You Load Your Car: Add "(if issued)" after Barcode scanner as was done for the Cellphone bullet. 9) As You Load Your Car: On 9/5/16 I had pointed out the misplacement of the "This Pocket Guide" bullet. Don't require the CJ PG to be in the blue plastic bag. See first CJ PG comment above. 10) How To Handle a Ballot Jam: "voters in the voting booths" in two places is not strictly accurate. For example, the voters may also be in line at the tabulator, but behind the person whose ballot caused the jam. The algorithm needs to count all of them: in booths, in line at the tabulator, waiting for a booth, walking from a booth toward the tabulator line, etc. -- all those people having a ballot still to be entered into the tabulator, but NOT the person who caused the jam and who has the ballot under consideration. This same flowchart appears in other places; see below. 11) Handling Media Communications: This chart will now need to consider the new rule that the 50-foot boundary also applies to the curbside area. Tabulator User Guide (undated) --------- ---- ----- Page 6: The Emergency section should get in sync with the General Election Manual page 8. For example, the manual directs the reader to the CJ PG, which says how to handle an unreadable ballot found in the emergency bin. The User Guide does not address this possibility directly, but does not punt to the CJ PG either. There may be other inconsistencies. Page 11: See "How To Handle a Ballot Jam" comment above. Voter Assistance Quick Guide (8/22/16) ----- ---------- ----- ----- Page 2: See "How To Handle a Ballot Jam" comment above. Help Table Quick Guide (8/22/16) ---- ----- ----- ----- Page 2: "Moved After 10/9/2016" block: This is analogous to my 9/5/16 comment on Flowchart #2. Using the old address may not be proper if the voter had long ago moved from that registered address and had lived somewhere else (maybe multiple places) for 30 days or more at each one before moving to the current address where he had lived for less than 30 days. Is it really correct to infer that his old (registered) address is the best one to use? Shouldn't the HT person or the Help Line try to determine the most recent 30-or-more-day residence address? 2016 General Election Manual (8/23/16, 10/16/16) ---- ------- -------- ------ 1) Pages 28, 76, 78, and 86: All are affected by the change to include the curbside area in the buffer zone 2) Page 52: See "How To Handle a Ballot Jam" comment above. 3) Pages 32 and 63, ID sections: One place says "fewer than 5%" and the other says "fewer than 2%" will have the ID designation on the ATV. I suspect the "2%" figure is more accurate, but both pages should have the same number. Suggestion ========== One of our officials, who had worked at early voting, told us of a procedure used there. When the curbside official got an "I Voted" sticker to take back to the voter, the official would write the current tabulator count on the sticker's paper backing so the voter could be told the tabulator count for that voter's ballot. Monday Pickup ====== ====== The "Stay in cars" edict did not work too well for Monday pickup at Knightdale. There had been no clear path set up for cars to line up. So the initial cars just parked near the van, waiting for the van people to start things at 4:30. But the van people later said they did not know those waiting people were there to do a pickup; in fact they said they were ready to deliver before 4:30. When they finally recognized the situation, they got out of the van and started the delivery to the CJ's who by then had gotten out of their cars. The van people then did the delivery very efficiently to people outside their cars, as they had for past elections. In contrast, "Stay in cars" worked fine for the Tuesday night delivery. Then there was a clear line where cars should wait. That had been the procedure used for Tuesday night in all the elections I had handled. It has worked for delivery; it's the pickup where "Stay in cars" doesn't work. See http://jgkhome.name/WakeBOE/BOE_BOE_Wish_List.txt for other suggestions. Report written by Jeff Knauth, Chief Judge for 19-09 with input from the other officials for this precinct