Controller 2014: Why the cost of a recount favors Betty Yee

The too-close-to-call race for state controller remained unchanged Friday morning — as Lake County has yet to count its 6,053 unprocessed ballots. San Mateo County, the only other county in the state with unprocessed ballots, has just four provisional ballots to review.

Board of Equalization member Betty Yee holds an 861-vote lead over Assemblyman John A. Perez, D-Los Angeles, for the final spot in the November run-off against Republican Ashley Swearengin.

What’s taking Lake County so long? According to Jim Miller of the Sacramento Bee, Diane Fridley, the Lake County registrar, “is on light duty following surgery and has only a skeleton staff to help with the ballot work.”

Average error rate: 1 in 6,400 votes

There’s been some speculation that the magic number is 645 votes. Recounts are rare, but the best available information comes from a study by The Center for Voting and Democracy, which analyzed statewide recounts from 2000-2009. According to the study, “In the five cases in which the total votes cast were above two million, the margin shift was on average 0.016% of the vote (fewer than one for every 6,400 votes cast).” Applying that rate to the current state controller’s race yields a potential 645 vote change.

But, there’s another piece of new data for political analysts to chew on: the recount in the 31st Congressional District.

San Bernardino County Error Rate: 1 of 1,278 votes

Third place Republican congressional candidate Lesli Gooch, who was just 209 votes behind Redlands Mayor Pete Aguilar, requested a recount in her Inland Empire-based race. Gooch was hoping to set up an all-Republican general election against first place finisher, Paul Chabot.

Gooch already has abandoned the recount in San Bernardino County. “Day one of the recount, which cost the Gooch campaign $6,330, tallied 1,278 votes and saw Gooch gain one vote on Aguilar,” writes Nikie Johnson of the Press Enterprise.

CD 31 applied to state controller’s race

That’s an error rate of 0.0782 percent (1 out of 1,278 votes), and if applied to the state controller’s race, would produce a statewide change of 3,156 votes. More than enough for Perez to make up ground on Yee. But, there’s no way to guarantee that the errors are in Perez’s favor. Yee is just as likely to see a recount add to her vote total.

To be clear, the sample size of CD 31 is incredibly small — too small to confidently project the error rate statewide. Again, the average margin shift from the study by the Center for Voting and Democracy, which analyzed more data, showed a change in 1 out of 6,400 votes cast.

Perhaps the more important number from the 31st Congressional District recount, which properly scales to the state controller’s race, is the cost of a recount. Gooch spent $6,330 to pick up one vote. If you apply that rate statewide, it would cost Perez $5.4 million to gain the 861 votes that he is currently down (861 votes x $6,330 = the cost to gain one vote).

Projection: Perez down as many as 434 votes

The latest projection published in Scott Lay’s The Nooner shows Perez down as many as 434 votes and up as many as 43 votes, once Lake County finishes counting. That would mean a recount cost of $2.74 million for Perez and $272,190 for Yee. As of the last campaign finance report, Perez had $1.8 million in cash on hand.

If our math’s right, that’s a very big number. California Democratic Party coffers are full, but that’s still money that could be used to pick up legislative seats in November. Even an outside benefactor, who could independently pay for a recount, would face criticism for spending heavily on the endeavor.

California State Controller Results, as of Friday June 27

Candidate	Party	Votes	Percent
Tammy D. Blair	Dem.	200,225	4.963%
John A. Pérez	Dem.	876,402	21.725%
Betty T. Yee	Dem.	877,263	21.746%
David Evans	GOP	848,846	21.042%
Ashley Swearengin	GOP	1,000,618	24.804%
Laura Wells	Green	230,755	5.7%