Wednesday, May 23, 2018

California Polling

There's a new USC Dornsife/LA Times poll out and I'm not going to link it. Why? Because it's such an embarrassing poll for USC that they should really quit polling. It's an entirely online poll. There are good reasons to do online polling but polls that were entirely online produced some really funky results in 2016. They aren't as accurate as polls that include phone calls.

The poll was conducted between April 18 and May 18. I've never heard of doing over a period of a month because the data is inconsistent. The people who answered in the early part of the poll could've made up their mind or even voted by the later part of the poll. Their answers are meaningless. That's why polls are conducted over 3-4 days.

This poll has roughly 40% undecideds in both the gubernatorial and senate race. That's really strange so close to the election and inconsistent with other recent polls. A poll with that many undecideds doesn't tell us how people will vote. The 2014 primary had 4.5 million votes. So far over 500,000 ballots have been returned. So 10-12% of the voters have not only made up their minds but their vote is set. At this point a pollster should have two samples, people who have voted and people who will vote. That'll give them an accurate idea where the election is.

California political data expert Paul Mitchell on why pollsters are so wrong on California polling. California has a ton of data on who votes. Yet pollsters ignore it. Mitchell shows the share of voters by partisanship, geography, and ethnicity. Latino turnout in primaries has hovered around 12%. Yet the pollsters are showing on average a 24% Latino electorate. They look at what share of registered voters Latinos are and decide that's who is going to vote.

California has party registration. So we know the share of the electorate each party had in the past and we can get a good idea of how that fits in this year by looking at how the VBM returns compare to previous years. Right now it's 44%D/33%R. That's slightly more Democratic than 2014, 44%D/36%R, but less than 2016, 49%D/34%R. Mitchell points out that Republican average 31% of the voters in the public polls but have been 38% of the voters in the last three gubernatorial primaries. Republican share could certainly be lower this year due to Republicans re-registering as NPP but that's a lot lower.

Getting party ID correct is vital in top two. The biggest primary question is whether two Democrats make top two in the senate and gubernatorial races. A poll with 30% Republicans may miss that.

Thursday, May 17, 2018

How Will the California Primary Translate?

Is this a Democratic year? Do the special elections tell us that or are they unique one offs? One good way to tell will be the California primaries. Since it’s top two and everyone can vote for anyone it mirrors the conditions in November. So California will be a good indicator.

In order to give us an idea of what the primary results will tell us we’ll first look at the California general elections.

There were 29 congressional districts where a Republican faced off with a Democrat in the general elections in 2012, 2014, and 2016. The other 24 districts either didn’t have a Republican or a Democrat in one of those three general elections. Keep in mind that these 33 districts aren’t representative of California as a whole, just the districts where the two parties faced off. There are some heavily Democratic districts in there but many of the other 24 are the most heavily Democratic. So Republicans did better here than they did statewide.

Democrats got 54.6% of the vote in these districts in 2012, 51.6% in 2014, and 57.2% in 2016. So 2014 was a Republican year, 2016 a Democratic year, and 2012 in the middle. I want to reiterate that these aren’t the statewide vote. Democrats got 62.0% of the congressional vote statewide in 2012, 57.7% in 2014, and 64.8% in 2016. These numbers shouldn’t be taken as an accurate reflection of the statewide share either, however, since each district didn’t have a Republican and a Democrat running. In 2016, for example, these totals represent 60 Democrats and only 44 Republicans.

There’s a twist when looking at the primary, however. Even though 2014 was a very Republican general election the primary was actually even more Republican. In 6 statewide races Democrats got 56.3% of the two party vote in the primary but actually did better in the general election, getting 57.5%.

On the other hand, Democrats got 70% of the two party Presidential vote in the 2016 primary but only 66% in the general election. Hillary Clinton blew out Donald Trump in the general election but the primary was even more Democratic.

There are 13 congressional districts where the margin of victory for one party was 10% or less in one of the four primary or general elections in 2014 and 2016. They are CA-3, 7, 9, 10, 16, 21, 24, 26, 31, 36, 39, 49, and 52. There are 2 other districts that weren’t that close in any of the four elections, CA-45 and 48. Those 2 are expected to be close this year.

In 2014 Republicans averaged a margin of victory of 13.6% in the primary in these 15 districts. That dropped to only 5.8% in the general election. In 2016 Democrats had an average margin of victory of 1.2% in the primary and 2.4% in the general election.

The statewide primaries for senator, governor, and other downballot races should tell us a lot about what kind of year we should expect. If Democrats are close to the 70% they got in 2016 we should see a very Democratic year. Even when they did get 70% they didn’t win any district that they didn’t get more votes in during the June primary. If it’s a heavily Democratic primary Democrats shouldn’t count on winning any district in November that they don’t win in June. The Democratic statewide vote share in June is closer to the 56% of 2014 there won’t be a blue wave, although we should expect Democrats to make gains on their totals in November.

There’s a huge middle ground here between 56% and 70% and chances are the average statewide vote will fall between 2014 and 2016. Is it possible for Democrats to get more than 70% in a statewide vote? I don’t think so. The last election was a perfect storm for them. Competitive Democratic primaries with uncompetitive Republican ones in a Democratic year. I can’t see that replicated.

We don’t know what the primary statewide vote would have to be for Democrats not to make much in the way of gains but we probably can tell how Democratic the year is based on where it falls between 56% and 70%.

Friday, May 4, 2018

Gavin Newsom's Strategy Could Backfire

California gubernatorial candidate Gavin Newsom is out with an ad linking John Cox to Donald Trump and saying he's too conservative for California.

The LA Times has deduced that Newsome is doing this to help Cox get conservative votes. That'll propel him into the top two and Newsom would be better off facing a Republican than a Democrat. This was done by Claire McCaskill to great success in the 2010 Missouri Senate race. These ads could certainly help Cox, siphoning off votes from his conservative rival Travis Allen. Mission accomplished! High fives all around!

Wait

There's a big difference in California. Unlike Missouri, California has no Republican primary, only top two. It'll be a race where everyone can vote. That includes independents and moderate Republicans. In the 2016 election there were voters who vote Republican but voted for Hillary Clinton. In CA-39 Ed Royce won 57%-43%. He won 58%-42% in 2012. That's pretty consistent. The district went from Mitt Romney winning 51%-47% to Hillary Clinton winning 51%-43%. In 2012, Royce overperformed Romney's margin by 12%. In 2016 he overperformed Trump by 22%.

It's possible that these once loyal Republican voters will vote Democratic in 2018. They've already done it once. Their reason for voting Democratic then was Donald Trump. That's the same Donald Trump that Newsom is tying Cox to. These ads could drive suburban voters from Cox to a Democratic alternative like the moderate Antonio Villaraigosa. While Cox may pick up Allen voters, he might lose some to Villaraigosa. Cox might end up with a net loss in the mix while Villaraigosa gains. Newsom could help get the result he's trying to prevent.

Oops

Friday, April 27, 2018

California IGS Poll: Bad News for Villaraigosa; Good News for de Leon?


California Berkeley IGS has put out a California primary poll based on online surveys. I'm not sure I've seen Berkeley's polling before but it has far better methodology than SurveyUSA and they actually polled all the candidates.

Governor
Newsom 30, Cox 18, Allen 16, Villaraigosa 9, Chiang 7

If this poll is accurate Villaraigosa has little chance. Cox and Allen split the GOP vote and still lap Villaraigosa. Villaraigosa is actually 5th with NPP voters, so undecided NPP voters should actually put him further behind when they decide. Unfortunately for him, Democratic voters have the lowest percentage of undecideds. Villaraigosa's only path to second place is to take a lot of Democratic votes from Newsom. It seems unlikely he could grab enough. Since the state is so heavily Democratic the perception is that only a Democrat could beat Gavin Newsom in the general election. I don't see any reason to disagree with that but it's worth noting that John Cox and Travis Allen are in a genuine close race for second now.

Senate
Feinstein 28, de Leon 11, Bradley 10

Wait. What? Who is the Republican leader James Bradley? Bradley's strong showing has perplexed the Sacrmaneto Bee and San Francisco Chronicle.

The answer is hardly complicated. First, let me debunk two theories I've already seen. The first is that the names were listed in alphabetical order and that Bradley benefitted from that. In Chronicle article Mark DiCamillo, the poll’s director, disputes this since “since the first five names listed were all Republicans and that only one of them received significant support.”

Secondly, there’s no evidence to support that Republicans will pick an anglo sounding name. In 2014, none of the Republicans were well known or ran much advertising. Yet Neel Kashkari beat Tim Donnelly by 4.6%. The Republican with the most votes in the 2016 Senate race was Duf Sundheim. I’ve never heard of the first name Duf before and have no idea the ethnic origin of Sundheim. Names like Phil Wyman and Greg Conlon are more Anglo sounding and yet Sundheim beat them.

To understand the answer you need to know how California voters pick their candidate when they have no information other than what's in the voting booth. The ballot gives them one snippet, ballot designation. That's the occupation listed below the candidate on the ballot. Ballot designation almost propelled unknown David Evans into top two in the 2014 Controller race.

James Bradley’s ballot designation is “Chief Financial Officer.” The other Republicans list occupations like "Teacher," "Bus Driver," and "Civil Rights Advocate." While those are all admirable professions they aren't what Republican voters are looking for in their candidates. They like ballot designations that show success in business. Only three of the Republicans have ballot designations that’d appeal to Republican voters and Bradley’s is probably the strongest. When I brought this up to California election data expert Paul Mitchell, he agreed with my assessment. So mystery solved.

The poll is still good news for Kevin de Leon even with so much of the Republican vote going to Bradley. de Leon is getting a decent 15% of Democrats and 15% of NPP voters. Dianne Feinstein is only getting 19% of NPP voters and all the Republicans combined are getting 12%. If undecideds break the way the voters in this poll do, de Leon gets 23% of Democrats and 28% of NPP. That'd give him around 18% of the vote and there's a decent shot he makes top two even if Bradley is as strong as he looks here.

Wednesday, April 18, 2018

Kevin de León Has No Path to Getting Elected

Today there's a story that Billionaire Tom Steyer is endorsing Kevin de León and could fund ads for him. People talk like de León has a chance but none of these people outline his path to victory. All you get is a vague answer about how other candidates have run to the left and won. Gavin Newsom is an example. This doesn't take into account two things. None of these candidates ran against a popular long time Democratic incumbent and none of them ran exclusively as a progressive champion. Kamala Harris ran as the attorney general who fought for you, not on progressive positions. I can't find Gavin Newsom ads on line. While he's staked out a position to the left he's not running primarily on that.

Running as a progressive champion may be all Kevin de León can do. Unlike Newsom and Harris people don't know who he is and he doesn't have a record like Harris. SurveyUSA had Feinstein ahead 52%-7% among Democrats and 22%-5% among independents. He isn’t competing for the 35-42% of the voters who lean right. An incumbent like Feinstein has a huge advantage due to her long record in the job and her popularity. There are also a percentage of Democrats who might be inclined to vote for the more progressive candidate but would support an incumbent over a challenger. It'd be disloyal to Feinstein to vote against her.

de León is trying to get enough of the voters who lean left and hoping for a fractured Republican field. It’s possible he gets that. Loretta Sanchez did, although she did so with no incumbent, high Democratic turnout, low Republican turnout, and running to Harris’ right, not her left. The general election is impossible. Republicans will vote heavily Feinstein, probably better than 80%-20%. One flank of de León's campaign is how awful Republicans are. He’s actively telling them he doesn’t want their votes and vote Feinstein. While some will leave the ballot blank, most will vote for one of the two candidates.California had 13.7% less votes in the 2016 senate race than the presidential race. Most other races were in 2% range. The two right leaning Presidential candidates got 35% of the California vote. So, at most, 1/3 of right leaning voters skipped the race. He’d have to win Democrats and independents by 25% to overcome that. And right now he’s losing Democrats by more than 7 to 1.

Thursday, March 22, 2018

PPIC California Poll

PPIC is out with a new California poll and it's a bad one for Antonio Villarigosa. He's now third behind John Cox. The underlying numbers aren't encouraging for him. 30% of Republicans, 26% of independents, and 17% of Democrats are undecided. Villarigosa is running 4th among independents, so he's likely to lose ground when those 77% of the undecideds decide. If Chiang or Easton get some traction he'll probably fall to 4th. One big problem he has is that Newsom even beats him with moderates. And he's running a far left campaign.

The Senate poll is junk. They include only two candidates when there are two dozen running. That's why 39% are undecided here, but only 24% are in the gubernatorial poll. Less Democrats are undecided in this poll but Republicans and independents sky rocket. De Leon gets killed by Feinstein among Democrats and independents but is close to her among Republicans who choose between the two. That isn't surprising even though de Leon is running on a progressive anti-Republican/Donald Trump message. I’m guessing that de Leon is angry that any Republicans would vote for him. These are people he despises.

Republicans don't know who he is, so some choose him over Feinstein. They know they don't want her. If Republicans actually hear about de Leon he wouldn’t do so well head to head. There are 4 Republicans in the race. I'm thinking Republicans will get 31-36% of the vote and Democrats 57-62%, with the remaining going to other parties. If de Leon is only able to get 15 % of the vote he needs a fairly even split over 3 or 4 candidates and hope one doesn't break out. Elizabeth Emken did in 2012, to some extent, but no one did in 2016. De Leon finishing top two is possible, although I think his chances are less than 50%. He has no path to victory in a one-on-one match-up. People registering Democratic aren't abandoning Feinstein and there aren't enough independents on her left for de Leon to beat her with them.

I keep hearing that no Republican will finish top two because there aren't any well known Republicans in the race. I guess this goes with theory that people will vote for whoever they know and/or see ads for. Um... No. Republicans will vote for a Republican they've never heard of instead of leaving the ballot blank or voting for a Democrat they dislike. A Republican Senate candidate could make top two without anyone knowing who he is.

Friday, January 12, 2018

SurveyUSA on California Governor and Senator

We've seen a few polls for the California gubernatorial and senate races but one problem common to all the polls was that they didn't actually survey the entire ballot. They chose the candidates they thought would do the best and polled them. This wasn't as big a problem for the gubernatorial race, as the polls included as many as six candidates, four Democrats and two Republicans. The senate polls, however, included only two Democrats, Dianne Feinstein and Kevin de Leon. And then they gave the shocking revelation that two Democrats would finish top two! Hmmm. You included only two candidates and both finished top two. What are the odds?

I've had a lot of criticism for SurveyUSA's polls. They've included voters who say they're going to vote and have ended up with 80% of the voters they've reached included in the survey. A primary might get a third of that. In this new poll they only include 55% of registered voters. Probably still too high but definitely better than previous polls. SurveyUSA includes everyone who'll be on the ballot and even some who probably won't be. They have over 30 candidates for governor and 20 for senator. It may be difficult for people to keep track of so many candidates in a phone survey but it's more reflective of what the actual ballot will be.

In the gubernatorial survey Gavin Newsom leads fellow Democrat Antonio Villaraigosa and Republican Travis Allen 18%-10%-9% in the gubernatorial poll. More than twice as many Republicans were undecided than Democrats. I gave the undecideds a choice based on what their fellow Republicans, Democrats, and independents said and ended up with Newsom 23%, Allen 15%, and Villaraigosa 12%. That'd put a Democrat and a Republican in top two and would mean an easy Newsom victory in November. Villaraigosa wouldn't make top two because of the number of Democrats in the field, but also because he isn't seen as an alternative for Republicans in the survey.

The senate poll produces very different results than previous surveys. In the Capitol Weekly poll that included only two Democrats Kevin De Leon beat Dianne Feinstein among Republicans 36%-17%. I theorized this was because Republicans had no idea who De Leon was but knew they didn't want Feinstein. When the four Republicans who are running are included on the ballot, those Republicans get 47% of the Republican vote, with Feinstein getting 6% and De Leon getting only 2%. As a result, De Leon finishes 6th, behind all four Republicans.

The survey isn't all bad news for De Leon, however. Feinstein beats him 58%-6% with Democrats and 25%-2% with independents. Once De Leon gets his name out there those margins should shrink, perhaps quite a bit. Secondly, the four Republicans are unknown and they split the Republican vote almost equally. If De Leon can improve his numbers with Democrats and left leaning independents he can make top two because the four Republicans split the vote. My projection is Feinstein 35%, De Leon 15%, and the Republicans in the 9-11% range.

The big difference between the gubernatorial and senate races is that Travis Allen is well known with the grassroots and has garnered some support. If Republicans don't run someone Republican voters have heard of, their candidates will split the vote. It's not a high bar for Republicans. They don't need a stellar candidate with huge name ID, but they do need someone voters can identify.