CA120: Issa down; Prop. 61, death penalty repeal up

ballot-propsWhat a year it has been for polling-related news in California – please try to contain your excitement.

The venerable Field Poll went online (shades of Dylan Goes Electric).

Meanwhile, USC and the LA Times combined to produce the most, um, “noteworthy” poll of the cycle (shades of Dewey Defeats Truman), which polling  Director Dan Schnur  posted on twitter “is wrong, but still gives us important info on Trump, Clinton voters.” His full discussion of the poll and its value, posted Monday, is available here.

With this poll, we just want to know who’s going to pick up champagne at 8:01pm when the first results come out, and who’s looking for Pepto-Bismol after that first report.

Polling methodology may not be exciting, but the results often are, and these results [pdf] are no exception. We’ll add more results later as we go through the numbers. But these are among the findings that are most likely to get politicos agitated or excited:

–We show Republican Congressman Darrell Issa trailing in CD 49.
–Challenger Ro Khanna is leading incumbent Rep. Mike Honda handily in CD 17.
–Josh Newman, who won an upset second place in the primary, is leading Ling Ling Chang among absentees in SD 29
— Al Muratsuchi looks to be sailing back into the Assembly after losing his seat in 2014.

Our earlier CA120 “exit” poll of California primary absentee ballot voters, released the evening before the June 7 election, received positive notice from major political data journalism sites.

“One of the coolest polls I’ve seen,” said Nate Cohn of the New York Times, followed by a number of “the election was rigged” exclamations from pro-Sanders Reddit users who claimed that the exit polling was more valid than the actual election results.

Our only intent was to see if we could predict what the first report of absentee votes would show.

But, more often than not, the fact that the absentees were such a high percentage of the primary electorate made the leaders in our poll the ultimate winners of the election.

And that brings us to the general election.

This is an internet-only, English-only poll.  Our sample is limited to potential respondents whose email addresses are on the voter file.

We have a reasonable demographic distribution naturally, and we weight our data to match the demographics of the expected absentee voting universe in the toplines here.

But we are principally relying on newer registrants and re-registrants to stand as proxies for those whose email addresses are not available.

The statewide trends mirror those we found in June – a weakness among Republicans.

Tomorrow will see a substantial poll -voting, late-absentee and provisional-vote population, so an early vote advantage can be overcome with turnout that is different in its partisan or demographic composition.

With this poll, we just want to know who’s going to pick up champagne at 8:01pm when the first results come out, and who’s looking for Pepto-Bismol after that first report

As we did in June, we have weighted our statewide sample to match the expected absentee voter universe using one of Political Data’s common target universes, and breaking that down by party registration, gender and age.

The raw unweighted data for every race, along with the breakdowns for candidate support among key subgroups will be made available tonight and distributed via the CA120 twitter account.  Additionally, we weighted key congressional and state legislative districts individually for analysis below.

The statewide trends mirror those we found in June – a weakness among Republicans.

The expected early-vote universe is 28% Republican.  However, when we asked respondents how they think of themselves politically, only 23% say they identify as Republicans.  Similarly, only 77% of the Republican early vote is for Donald Trump, while Hillary Clinton is supported by 93% of Democrats.

Non-partisans (voters not registered with either major party) break towards Clinton by better than three-to-one (60% for Clinton and 17% for Trump).  This is key because down-ballot Republicans will enjoy better party unity than Trump, but can still be overwhelmed by non-partisans who follow their Presidential vote.

A few incumbent congressional and state legislative Republicans in and around Orange County (Inland Empire and San Diego) will have a very big pit in their stomachs tomorrow night.

In the 2012 CNN exit poll, Barack Obama defeated Mitt Romney 55%-39% among California Independents (based on self-identification), so clearly Donald Trump will lag that total significantly.  To the extent those voting patterns hold, any Republican in a district with a GOP registration advantage that is only in the single digits will have a very close race.

I expect that a few incumbent congressional and state legislative Republicans in and around Orange County (Inland Empire and San Diego) will have a very big pit in their stomachs tomorrow night – and the biggest regret among Democratic strategists is that, at least on the congressional level, they didn’t target more seats.

This tracks our June poll that accurately found much stronger Democratic performance in this area than past history suggested.  Wouldn’t be funny if an orange-hued man was responsible for turning Orange County blue?   I wouldn’t bet against it.

There’s little drama statewide.

Hillary Clinton will have more than a two-to-one advantage in the early vote as discussed above.  In the U.S. Senate race, Kamala Harris will have a massive lead as well over Loretta Sanchez.  Sanchez’s plan to court Republicans has not been successful – she trails Harris narrowly even among early GOP voters.

Down-ballot, we see the challenges facing Republicans as they fight to remain relevant.

We have Clinton ahead of Trump in the early vote in CA 49 by 20 points and Applegate ahead by 16.

In the only Congressional district where a Democratic incumbent was seriously challenged (CA 7 in the Sacramento area), Democratic Rep. Ami Bera holds an 18-point lead in the early vote over his Republican opponent Sheriff Scott Jones.  The Congressional vote matches the presidential vote closely – Bera receives 58% of the early vote to Clinton’s 54% while Jones at 38% does not significantly exceed Trump’s 32%.

The likeliest Republican incumbent to survive a well-funded, well-organized Democratic challenger is Catherine Baker in the 16^th Assembly District.  In the most expensive Assembly race in the state, Baker holds a three-point advantage in early vote over Democrat Cheryl Cook-Kallio.  It is striking that Baker can have an advantage in a district where Clinton has a three-to-one advantage (66%-22%).  That trend shows real cross-over strength for Baker, but could still make for a rough road ahead with the Election Day voters.

In June, we reported that Rep. Darrell Issa of the 49^th Congressional District was in trouble when no one saw it coming.

Our early vote exit poll had Issa tied with his unknown Democratic challenger Doug Applegate and in the final results, Issa had 51% to Applegate’s 46%.  With an influx of resources on behalf of Applegate, a more Democratic-leaning general electorate, and most importantly a massive break among non-partisans in this district toward Democrats we expect Applegate to win the early vote by a significant margin.  We have Clinton ahead of Trump in the early vote in CA 49 by 20 points and Applegate ahead by 16.

In the 25^th Congressional District, we find Democrat Bryan Caforio tied (45%-44%) in the early vote with first-term incumbent Republican Rep. Steve Knight.

Democrats have been frustrated by their inability to win in the Central Valley for many years.  Entering Election Day, they may be able to finally pull out a victory.  In the 10^th Congressional District we show Democratic challenger Michael Eggman leading Rep. Jeff Denham in the early vote 51%-44%.  Clinton leads Trump by an identical seven-point margin.  Denham enjoys slightly better party unity than Eggman and Eggman’s advantage among non-partisans is just a few points.  Unless the composition of Tuesday’s electorate varies greatly from the early vote, this should be a close result.

More recently, Democrats have been looking to break through in the Antelope Valley/Santa Clarita portion of Los Angeles County.  In the 25^th Congressional District, we find Democrat Bryan Caforio tied (45%-44%) in the early vote with first-term incumbent Republican Rep. Steve Knight.  Knight may be better positioned to survive than some of his colleagues because Caforio is underperforming the top of the ticket.  Clinton leads early vote in this district 50%-38% and Knight enjoys better party unity than Caforio.

 We find challenger Ro Khanna well ahead of Rep. Mike Honda in the 17^th Congressional District.

Those trends are even more pronounced in the open state Senate race (SD 21) in that region.  Republican Scott Wilk has a small advantage in the early vote over Democrat Johnathon Ervin (44%-40%), despite Clinton holding a nine-point advantage. Because Election Day turnout tends to favor Democrats, both races are winnable for Democrats, particularly Caforio’s, but the lack of party unity among Democrats is a big concern.

Outside of that portion of Los Angeles County, the remainder of the state’s most populous county gives insurmountable early vote advantages to Democrats in contested races; most notably SD 25 (Anthony Portantino), SD 27 (Henry Stern) and AD 66 (Al Muratsuchi).

One of the surprise results we identified in June was in Senate District 29, which is mostly in Orange County.

We found the two Democratic candidates’ combined vote equal to that of the Republicans when past history indicated a fairly substantial GOP margin was expected.  Also, we found that then-unknown Democrat Josh Newman had a slight advantage over the better-funded, better-connected Sukhee Kang.  Newman finished ahead of Kang and now has a slight advantage in early vote over Republican Ling Ling Chang (46%-40%).

On the statewide measures, this early-vote universe identifies as more liberal than conservative by 16 points.

As in the Antelope Valley races, Newman is trailing the Clinton margin (52%-36%) pretty substantially, and that is the challenge.  This should be a tight finish with a result we might not know for weeks.

In some of the higher-profile Democrat-vs-Democrat general elections, we find challenger Ro Khanna well ahead of Rep. Mike Honda in the 17^th Congressional District, and state Sen. Jim Beall, Jr. should have a comfortable early vote advantage over Nora Campos.

In the open 14^th Assembly District, Democrat Tim Grayson opens up with an 11-point advantage over fellow Democrat Mae Torlakson.  It is noteworthy that Grayson is close to Torlakson – the wife of popular Democratic Superintendent of Public Instruction Tom Torlakson – among Democrats in the early vote.  However, more base Democratic turnout probably causes this race to tighten tomorrow.

On the statewide measures, this early-vote universe identifies as more liberal than conservative by 16 points.

That means the liberal side of high-profile ballot propositions have big advantages in this survey: Yes on Proposition 56 ( tobacco tax) is ahead by 40 points, Yes on 64 (marijuana legalization) is up by 34, Yes on 62 (ending the death Penalty) is ahead by 10,  Yes on Proposition 61 (prescription drug prices) is ahead by 10, though a much larger “don’t recall” response was found here compared to other measures.

For those who want to see the raw, unweighted data in our surveys, click here for Congress, here for state Senate, here for Assembly and here for the presidential.
—
Ed’s Note: Includes full comment of Twitter posting and adds link to discussion of the poll in the LA Tmes, 3rd graf. Pollster Jonathan Brown, a regular contributor to Capitol Weekly’s CA120 column, is the president of Sextant Strategies and Research. Poll administration was handled by Paul Mitchell of Political Data Inc., with assistance from Alan Yan of UC Berkeley’s Political Science Department.