Man or machine? Rossi-Gregoire recount prompts dispute over accuracy

Associated Press ^ | 11 December 2004 | David Ammons

[WKL815]

I observed the Kitsap County recount.

I happened to watch a table of two older gentlemen, late 60's and 70's. In addition to the Republican one setting a Rossi vote into the overcount pile because he thought two light stray marks making an "x" in the write-in and libertarian arrows qualified it as such, the two men had many problems counting the ballots and batching them in groups of 25 as they were supposed to do. (I got the Rossi vote out of the overvote pile after talking with our county auditor)

Each party rep is supposed to count the other party's ballots first and batch them into groups of 25. Well the Republican party rep kept under batching the Gregoire votes creating a superficially increased count for her. The Democrat consistently overbatched the Rossi votes creating a superficially decreased count for Rossi. Now whenever the numbers they report don't match the machine count, the table's election official has to come back to the table and tell the two they need to count again. So the two switch and count their own party's ballots. This is where I observed the numbers changing and realized the batching problem. Towards the end of the day, the Dem had one precinct that he mistakenly hid 8 votes for Rossi. Now I don't think there was fraud intent there, just human error in a repetitive task hard for older hands to do.

But I certainly see now where there's room for human error and boy do we need observers and counters who know they are there to protect votes.

[WKL815]

I observed the Kitsap County recount.

I happened to watch a table of two older gentlemen, late 60's and 70's. In addition to the Republican one setting a Rossi vote into the overcount pile because he thought two light stray marks making an "x" in the write-in and libertarian arrows qualified it as such, the two men had many problems counting the ballots and batching them in groups of 25 as they were supposed to do. (I got the Rossi vote out of the overvote pile after talking with our county auditor)

Each party rep is supposed to count the other party's ballots first and batch them into groups of 25. Well the Republican party rep kept under batching the Gregoire votes creating a superficially increased count for her. The Democrat consistently overbatched the Rossi votes creating a superficially decreased count for Rossi. Now whenever the numbers they report don't match the machine count, the table's election official has to come back to the table and tell the two they need to count again. So the two switch and count their own party's ballots. This is where I observed the numbers changing and realized the batching problem. Towards the end of the day, the Dem had one precinct that he mistakenly hid 8 votes for Rossi. Now I don't think there was fraud intent there, just human error in a repetitive task hard for older hands to do.

But I certainly see now where there's room for human error and boy do we need observers and counters who know they are there to protect votes.

[JosefK]

Thanks for posting. I really need you to help me ensure that GOVERNOR-ELECT Dino Rossi stays GOVERNOR-ELECT DINO ROSSI and please call 425.646.7202 ASAP!

[Veto!]

By the way, "somebody" turned in my Democrat partner, and she was fired.

Good going, Recount Warrior! Applause, applause. You're REALLY somebody!

[supercat]

The only way I would trust a hand count would be if they gave different teams the same group of ballots and certify the count only after the teams independently reach the same tally.

The only way I would trust a handcount would be if the ballots were run through a laminating and serializing machine (device to mark each one with a serialized barcode) and two independent groups of examiners were to record all the the ballots, and if the two individual groups recorded the individual ballots identically.