2004 Ballot Recount: Observer Report

• December 15, 2004:
  Report by Green Party Observer

December 15, 2004:
Report by Green Party Observer

Upon arrival, I noted that the top floor of the building was shut down. Signs said the office was closed. If you did not know the recount was scheduled for that day, there would have been no way of knowing from the main entrance signs. It was only by luck that I tried the basement door and discovered where those involved were congregating.

A test deck was initially run, but it took some convincing to get the staff to actually do a hand count of the test deck to confirm the tabulator's validity. The tables with the recount staff lined one room. The observers were placed in chairs at least five feet away, while excess observers, the press, and the Board of Directors hovered back against the wall in the same room, watching the action from a distance.

Only one out of three counting stations was performing the counting process in a manner that actually allowed observers to see the punchcard ballots as they were counted. The other two teams did not bother to hold the ballots up for the observers to see. We challenged this procedure and were told that after the initial count, they would go back through the ballots, giving us the opportunity to get a closer view of the back sides to look for hanging chads.

Upon finishing the first run through, the counters flipped through the stack like a deck of cards, claiming this was the view of the back of the cards that we had requested. It was impossible to see the backs of the cards and many were missed in the process. Since one team had counted correctly, since the process had already taken several hours, and since no one questioned the integrity of the paired Democrat and Republican staff counters, the initial hand recount was accepted.

The machine count, performed by a Triad technician named Brett Rapp (who turned out to be the President of the TriadGSI company), found only one of the three precincts to have a hand count that matched to the machine tabulation.

The other two precincts were both off and had to be recounted by hand again. One team matched on the second try, the other took three tries.

Specifically, precinct 224 showed 333 votes for Bush in the hand recount and 334 votes for Bush in the machine count of the same sample.

A second hand recount and machine tabulation showed 334. This was blamed on inherent defects in the punchcards and the fragile nature of chads, which sometimes come loose and fall out in the machine tabulation.

We were not allowed to view the poll books or signature books. Just before 1:00 PM, a call came in from the Secretary of State's office informing the Board of Elections that we could view only the poll books and signature books from the precincts counted that day. However, the total 100 percent machine tabulation did not end until the regular closing time for the office and so we were not permitted to view any poll books or signature books and were told to return the following day.

The selection of the precincts was not at all random. Instead, three were chosen (31, 224, and 442), one for each counting team, that were approximately the same size and that were the right size to add up to the required 3% minimum for hand recount. It was pointed out by those familiar with the population of the precincts that the choices seemed to be very calculated based on ease of recount.

According to the Board of Elections officials, the tabulating machines are networked to the computers in the same room and all computers in the Board of Elections office are networked only to each other, not an outside line. Is this true? It seemed far too difficult to get a straight answer.

The gathering of additional data and canvassing of rejected provisional voters is still ongoing in Greene County. Supplemental reports will be forthcoming, provided by other witnesses present on the day of the recount.

Information for four rejected provisional ballots was provided for precincts in recount only for Precinct 442. There were 3 absentee ballots rejected for being received after November 2 in Prec. 442.

Additional Data for Precincts Chosen

Ballot counts for precincts chosen for Recount (31, 224, 442):