Friday, July 25, 2014

CA-25: Democrats Will Host a Debate

Two Republicans will be competing for the CA-25 congressional seat. This isn't unique. We had a number of these in 2012 and there'll be 23 single party elections in November. Some of those elections are between an established incumbent and a candidate who is just a name on the ballot. Others, however, promise to be competitive. In those, voters from the other party will likely be consequential. That's even in a district where they are a smaller percentage of the voters. Republicans were only 23% of the registered voters in CA-15 in 2012, but they were still the difference in Eric Swalwell's 52%-48% win.

Any candidate needs to find a way to appeal to people from the other party without abandoning their own party. These voters can make a difference, but you still need to get a substantial number of your own voters. So abandoning your positions to pander to the other party won't work. And that doesn't seem to be the strategy used in single party elections in districts like CA-8, CA-30, and CA-31 in 2012.

Democrats aren't inconsequential in CA-25. In 2012, the Democrat got 45% of the vote. Even with lower Democratic mid-term turnout and a 20% blank ballot, you're still talking 33-35% of voters being Democratic. If one candidate gets 80% of those voters, he only needs to get 35% of the remaining vote. That's fairly easy for a Republican who has an established base.

Since neither Tony Strickland nor Steve Knight can afford to get trounced with Democrats, they can't ignore them. Democrats know this and have invited the candidates to a forum on August 1. I'm sure neither wants to be there but if one goes the other can't afford to not go. I imagine this forum will be beating up Republicans with slanted questions and Barack Obama style straw men.

"Trickle down economics benefits the rich and hurts everyone else. Will you be supporting it?"
"Do you plan to work with or against the President?"

Obviously neither candidate can abandon their positions and start supporting Democratic positions. That wasn't necessary in any 2012 election. But how do you answer these questions in a way that appeals to Democrats but doesn't turn off your core voters?

Thursday, July 24, 2014

CA-52: Democrats Put Out Unbelievable Poll with Scott Peters Ahead

A big surprise in the California June primary was how badly Democratic congressman Scott Peters did. In 2012, Democrats lost the CA-52 primary by 3.5%. Scott Peters had a narrow 2.4% win in November. Peters entered this primary as an incumbent. So he would've been expected to do as well as he (and other Democrats) did last year. Instead, he lost to the Republicans by 15.4%. That's a lot to overcome and even would've been in 2012. Some people started to write Peters' political obituary. A little after the primary, KGTV-TV San Diego commissioned SurveyUSA to do a poll shortly after the primary and their results confirmed the primary result. They had DeMaio up by 7%. That's a terrible narrative for an incumbent, especially this far from election day.

House Majority PAC, a Democratic SuperPAC, released a poll that contradicts the primary and independent poll results. They have Peters leading DeMaio by 5%. The party ID breakdown in the poll is Republicans +8, a believable breakdown for the district. What's not believable is Peters taking 16% of the Republican vote and winning DTS/other by 24% isn't. In the SurveyUSA poll, DeMaio won independents by 20%. Republicans likely won independents by more than that in the primary. I don't find the survey results believable but I do find them understandable. After Peters bombed in the primary and SurveyUSA had him down 7 he needed a poll saying he's winning. Here it is.

Friday, July 18, 2014

California Controller Recount Over

Assembly Speaker John Perez called off the recount today after spending R30k this week and ending up picking up only 10 votes. I was surprised that Perez didn't pick up more than that because he was picking the precincts to recount. I actually predicted that he'd eventually go ahead. As it became apparent that he wasn't gaining votes in precincts that were advantageous to him, Perez calling off the recount was inevitable.

Perez lost the race by .01% and conventional election wisdom considers that any race that's within 0.50% is worth a recount. And Perez may have won in a statewide recount because the race was that close. There is no mechanism for a statewide recount. So Perez was on the hook for paying for the recount and he certainly couldn't afford to pay for it statewide. Look at how much cash he burned through just paying for select precincts in two counties. This has caused people to call for election law changes to have an automatic recount paid for by the state with an election this close.

So Democrat Betty Yee advances to face Republican Ashley Swearengin. Yee is a decent, but unspectacular candidate, whose strength is in the Bay area. I know some people think that Democrats winning a statewide office is automatic but Republicans have won a statewide office in every election except 2002 and 2010. In both those years the GOP lost races by less than 1%. So it'd be unprecedented if there isn't a race that Republicans at least finish less than 1% behind the Democratic victor. This race would seem like the best GOP hope.

Wednesday, July 16, 2014

Six Californias and the U.S. Senate

The latest news on Six Californias is that the proponents have turned in enough signatures and it may qualify for the 2016 ballot. This has caused a renewed look at the electoral implications. Nate Cohn of The New York Times takes a stab at an evaluation. While his analysis of the Presidential implications is solid, he doesn't really take much of a stab at the senate. He does write:

The implications for the Senate are obvious: 10 new Senate seats, four of which would almost certainly go to Democrats. The other six would be competitive, especially in presidential election years. If Democrats and Republicans were to split the six competitive seats, Democrats would net four seats in the Senate.

Cohn has fallen pray to the same thing many national pundits do, judging California elections based on Presidential numbers. The two aren't as correlated as they think. I looked at this in April, with the Feinstein-Emken numbers. Dianne Feinstein was an enormously popular sitting senator with a large war chest running against an unknown Republican with no money in a year where Barack Obama got a record Presidential vote. Feinstein won by the largest margin any Democrat has won by since 1986. So it's a high water mark for Democrats. Feinstein didn't win either Central California or Jefferson and if Feinstein isn't winning either in a good Democratic year, no Democrat is. She narrowly won South California, so that's the only new state they'd have any shot at. If they won that seat in a Presidential year, Democrats would get 7 of the 12 seats, no net gain over what they have now.

As I said, those Feinstein numbers are high water marks. While there are no elections in Six Californias now, there is the Board of Equalization. That divides California equally four ways and in a very helpful way. First, BoE 4 almost exactly mirrors South California. The only difference is that South California has all of San Bernardino county while BoE 4 has about half of San Bernardino county. While we don't have general election results for this district, we do have primary results. The Democratic party didn't bother to field a candidate, although one Democratic "some dude" did run. Republicans won the district 66%-34%. Democrats usually do better in November, but any district the GOP won by 32% isn't going to be competitive. Now it's possible that in a Presidential general election year with an actual well-funded candidate Democrats will do better, but they've generally done poorly in Board of Supervisor races in these counties. Their electoral success has come when the counties are carved up and heavily Hispanic districts have been created. That won't happen in a statewide election. Winning a senate seat would be an uphill battle for any Democrat.

The new state of Central California would be entirely in Board of Equalization district #1. The parts of BoE #1 that wouldn't be in this state are the more Democratic parts, mainly Sacramento County. Jefferson is in both districts #1 and #2. As with BoE #4, the Democratic party isn't contesting BoE #1. They again had a "some dude" who decided on his own to run. He lost 60%-40% in the primary. That's not as bad as BoE #4 but if we only take the part of BoE #1 that's in Central California, we get a 63%-37% Republican win. So a U.S. Senate seat looks like a longshot.

Jefferson is in BoE #1 and #2. Republicans didn't run a serious candidate in BoE #2, just like Democrats didn't field a serious candidate in BoE #1. So it's a fairly equal trade off. The Republican candidates in BoE #1 and #2 won Jefferson 53%-47%. That doesn't seem as bad for Democrats and it's a gap they could easily close in a general election. The problem with that is the candidates. The Democratic candidate in BoE #2 is from San Francisco, an area of strength in that seat. They don't have the same strong candidates in the Jefferson portion of the district, while Republicans have a strong bench in the Republican parts of Jefferson. I wouldn't be optimistic Democrats would win a senate seat here either.

The most likely result for Six Californias would be 6 Democratic and 6 Republican senators, with Democrats having a shot to win an additional seat.

Controller Recount isn't Working for John Perez

Yesterday was day three of the California State Controller election recount. While Imperial County numbers are unavailable, Kern county numbers are. They counted ballots where John Perez beat Betty Yee 1,652 to 521. That sounds like fertile ground for missed Perez votes. It wasn't. After the recount, those ballots were Perez 1,651 to 521. That's right. He lost a vote to Ashley Swearingen. Recounts usually produce a change in votes, but this one isn't working for John Perez. Either Perez needs to have the elections' boards count a lot more ballots or he needs to just call it a day and concede.

Tuesday, July 15, 2014

California Controller Recount: Perez picks up 4 votes

Two days into a recount that John Perez lost by 481 votes, he's picked up 4 votes. They're recounting his best precincts in the state and he's picking up nothing. That doesn't bode well for him to change the final result. If Perez doesn't start picking up votes soon, he's going to have to conclude he won't and stop the recount.

Sunday, July 6, 2014

The California Controller Recount is On

The California Controller recount is on like Donkey Kong. Former Assembly speaker John Pérez has requested a recount in 15 counties: Kern; Imperial; San Bernardino; Fresno; San Mateo; Orange; Ventura; Los Angeles; Riverside; Stanislaus; Tulare; Napa; Kings; Lake; and Merced counties. I haven't reviewed all of them, but none are small and they look like counties Pérez beat Yee in. So he figures to harvest a good amount of votes. I assumes he's got precinct level data and is only requesting certain precincts. That'll make a faster less expensive count.

Some of the counties are Republican strongholds. Swearengin and Evans got 65% of the vote in Fresno and Kern counties and one assumes that they topped 80% in the more rural areas. There won't be votes for Pérez there. All he'd be doing is helping the Republicans pick up votes. On the other hand, there are precincts in Bakersfield and Fresno he probably killed in. The number of votes is so small and the number of ballots they'll recount is so large that Pérez will almost certainly pick up enough votes to go ahead.

That doesn't mean that I think Pérez actually got more votes than Yee. But if they only count his strong counties and not hers, he picks up votes and she doesn't. The question is how Yee counters this. The standard is you only call for a recount if you're behind. But there may be a deadline for requesting a recount and the minute that Perez goes ahead, he'll stop paying for the recount. Can Yee wait until then and then start paying for a recount in San Francisco, Monterey, Santa Cruz, Alameda, and Marin counties?

Wednesday, July 2, 2014

Where John Perez should look for a recount

There is no mechanism for a statewide recount. Instead, John Perez (or any voter actually) has to request a recount in any county he wants recounted. He can also choose the precincts he wants recounted. So Perez would likely start with Imperial County, where he beat Betty Yee by a remarkable 31%. Unfortunately for him, Imperial is probably too small to entirely make up a deficit. He'd probably also need to select precincts in some Southern California counties that he beat her by 10% or more.

Since Perez is from LA county you might think he'd look there first. He beat her in LA County by only 4.8%, however. When it's that small, there's no guarantee he'd pick up votes in the county as a whole. It's less than the 8.0% he beat her in San Bernardino or 9.7% in Kern. He'll likely have to drill down and find LA county precincts that he beat her by 10%+. I don't have precinct data, so i can only assume such precincts exist. If there's enough of them in LA County, then he should go there first. One problem he has is that he comes from a low turnout assembly district. There were only 17k assembly votes in the district in the primary. There were 37k in the San Bernardino county AD-40, for example. So there aren't likely that many votes in his own district.