Replies

That is a terribly organized paper. But I went through part of it and this is what I believe it’s saying. Can someone critique this????

The looked at two counties. General and then some local election. Im still trying to figure out how this could get so screwed up. There are two main problems.

Mesa County

October 1
Database is set up/created. This includes a tabulation DB and a adjudication DB.
October 19
Ballot Scanning begins.

(*Problem #1)
From record or system timestamps in the DB, from 12:07:44 PM to 12:08:28 PM, 10 batches with 941 ballots are processed. A world record. The actual time to process, based on the automation rate, should have been 564 seconds.
No one knew or understood that the records, the entry, when the record hit the DB, a timestamp (date time) was inserted showing when the record hit.

267 batches with 25,913 ballots are processed until October 21.

October 21
2:18 PM
Another database is set up/created. This includes a tabulation DB and an adjudication DB.

(Problem #2)
Now there are TWO databases
This is believed to happen when someone stops and restarts the adjudication. At least that’s what the document suggests.

October 21
2:30 PM

209 batches are copied into the “new” database. They know they were not rescanned because of the super short duration (4 seconds each) which is impossible for physically processed. Thats not important, but this is....

(Problem #3) this is a big problem...

it appears some (batches 2-59) of the previously adjudicated ballots never made it to the new database.

Officials began to have issues. Someone notices something is off.

October 21
2:34 PM
Mesa County election officials start seeing the same ballots that were adjudicated before.

October 21
4:00 PM

Dominion contacted
Person from Dominion said they didn’t have a solution to the problem. BUT, the issue went away. No one knew what happened.

So they had to redo some of the batches. They know some 209 batches were processed twice.

HERE IS THE BIGGER PROBLEM. Same set of 209 batches:

First pass through, 2,166 ballots were tagged (by Dominion) for manual adjudication.

The second time through, 965 ballots were tagged (by Dominion) for manual adjudication.

WTF is the difference? No one can explain that. It’s the same set of batches, only one has less flagged now for adjudication.

This accounts for 25% of the ballots for the county.

Are you ready for this BS? I didn’t do through it all yet. but looking at the next example county.

Grand Junction Colorado.

To summarize, SAME THING. Second DB created, and the number of adjudicated ballots changed when processed a second time. 339 the fist time, 39 the second time once copied into the “second” DB. (see page 23 at the link)

That doesn’t make sense. Programs just dont randomly decide what to reject and what not to. Something is really off.