Sherril Huff's ongoing failure to properly account for ballots

Sherril Huff launched her campaign for Elections Director with this claim:

Under Huff's leadership, King County Elections received a national award for its mail ballot reconciliation efforts in 2006. Huff's team used Fortune 500 quality assurance standards, which revolutionized elections to performance and accountability standards similar to other successful businesses.

Baloney. Two months after receiving the so-called "award", Huff administered the September 2006 primary. Award or not, Huff gave us a recount with an unexplained error rate even greater than that of the 2004 recounts.

The first count in September 2006 ended with a close finish in the Northeast District Judge race between 2nd and 3rd place finishers Mary Ann Ottinger and Frank LaSalata. A recount was held to determine who would proceed to face Richard Pope in the general election.

King County counted 92,992 ballots in the first count, and 92,930 ballots in the recount. This is not a matter of reinterpreting the voters' marks. The number of actual physical pieces of paper changed. Either a net 62 ballots disappeared from the counting area during the recount, or those same 62 ballots never existed but were doubly counted the first time around. Percentagewise, that's a greater fluctuation in the ballot inventory than in the 2004 recounts, where a net 395 unexplained ballots seemed to materialize out of nowhere. Where was the award-winning ballot accounting in 2006? Either they counted as many ballots as there were voters in the first count, or they counted as many ballots as there were voters in the second count. But obviously not in both counts.

And while 62 ballots might not sound like a large number, LaSalata's final margin of victory in the race was only 27 votes. LaSalata went on to win the general and is now a sitting judge.

Huff's "explanation" for the accounting failure was sheer nonsense:

Visual inspection could have provided a fluctuation in write-ins, vote totals or other variances.

"A small amount of discrepancy between counts is not uncommon. At the end of the day we're still relying on humans to perform these functions regardless if the count is done by machine or manually," said Sherril Huff Menees, Assistant Director of Records, Elections and Licensing Services.

It is certainly possible that the recount produced the correct number of ballots. But what does that say about the "award-winning reconciliation" that was used to certify the first count? In fact, the DIMS records show that 93,010 were credited in that district, a net 18 more absentee voters than absentee ballots in the first count. That's quite plausible, given that in every election a small number of envelopes are inevitably returned empty and a small number of ballots are spoiled. I would tend to suspect that ballots were undercounted in the recount. But what's the point of having a recount if you have no idea how many ballots you have on hand in the first place? And how can anybody trust the result of a recount when the margin of victory is a fraction of the unexplained discrepancies? I suspect the only reason this story didn't get larger play at the time was because LaSalata's opponents were both somewhat eccentric and didn't have any institutional support behind them.

Even more serious than any errors that occurred was Sherril Huff's non-explanation for them. Either she had no idea why her staff didn't know how many ballots they had, or she knew, and covered it up and lied about it. Either way, Huff's campaign claim of "Fortune 500 quality ballot accounting" is a lie and a fraud.

Oh, and the number of ballots also changed in the only other recount that Huff has administered in King County, in November 2007.

Posted by Stefan Sharkansky at January 26, 2009 09:55 AM | Email This