Thursday, August 30, 2012

Democratic Internal Polls

There have been a plethora of Democratic internal polls released both nationally and in California. We've had polls on CA-41, CA-9, as well as CA-26, CA-47, and CA-52. We can add CA-7 and CA-3 to that list.

There have been a couple of Republican polls. Campaigns put out internals to show the race being closer than it is if they're the underdog or a bigger lead if they're the favorite. Polls can easily be skewed by the questions or by reweighting based on expected turn-out.

It's a sign that things aren't rosy for the Democratic party that there have been a lot more Democratic polls than Republican ones. On the other hand, Democrats have led or tied in all the polls. When a campaign releases a poll showing them "as close as" 9 point, throw dirt on them.

I'm skeptical Bera is really in this race, although I wouldn't dismiss him. Garamendi is likely doing well, but not that well.

Tuesday, August 21, 2012

CA-41: Mark Takano's internal poll

Mark Takano has put out an internal poll and it's hard to see it as too encouraging for the Democrat. Takano is up by 4 points. On the surface that doesn't seem bad. He's winning and Republicans did beat Democrats in June by 9 points here.

But we have to remember that it is a Democratic internal poll and this is a district that Barack Obama won 59%-38% and Jerry Brown won 51%-40%. So it's a district where a good Democratic candidate should win comfortably. Keep in mind that this is his poll, put out to show how well he's doing. Internal polls usually are skewed for the candidate conducting them, so he probably isn't really leading.

What's worse is that the poll has Republican Riverside Supervisor John Tavaglione with a 15% name ID advantage. If this were August 2011 I could understand a 15% name ID advantage, as Takano was largely unknown. But we're less than 3 months before the election. They should know him by now. What has Mark Takano been doing for the last year?

The popular meme is that Republicans are fading away and Democrats are dominant. As the primary results in CA-21 and CA-31 showed, however, the Democratic party has a number of places where they just don't win elections. Riverside county is one of those places. No matter how many votes Barack Obama or Jerry Brown get, you win locally with local candidates and the local party. The Democrats have no bench in Riverside county and no idea how to win elections.

Any decent Democratic candidate would win this district, but Takano isn't a decent candidate. He could win, but that'll be due to the overwhelming number of Democrats here.

Sunday, August 19, 2012

CA-15: Pete Stark Makes Headlines

I do enjoy Pete "what is he going to say next" Stark. Fortunately, there's a website devoted to keeping up. We've had Stark's insane berating of former Assembly Majority Floor Leader Alberto Torrico. Stark is congress' only self-proclaimed atheist, but decided to lecture Torrico on what it means to be a Christian. Which he thinks is a load of hogwash.

This led to Don Perata, who was President Pro Tem of the California State Senate to endorse Democratic Eric Swalwell in the 15th Congressional District race.

Even the liberal San Francisco Chronicle has abandoned him. I'm not sure how anyone thinks he'll win. Like Brad Sherman and Howard Berman, Stark is in a district that is 48% Democratic/52% other. Swalwell will get Democratic votes. There are some out there who find Stark an embarrassment and a dinosaur. I can't imagine he excites young progressives. Then the question will be how many Republican and independent votes will Stark get. If Stark gets 70% of Democrats, but only 5% of Republicans, he'll have to get 55% of independents to win. I just don't see it.

Saturday, August 18, 2012

CA-30: When Liberals fight

It's always amazingly incongruous when two rich liberals argue who connects with the little guy more. The Howard Berman-Brad Sherman fight continues to be ugly. I've wondered for years how Los Angeles Democrats would react to a genuine challenge and now we're being punished for that wish. Sorry. The district is around 49% Democratic and 51% Republican/Independent. Top Two was designed with this sort of race in mind to put the more moderate candidate in office. Neither of these guys are moderate and they certainly aren't pretending to be.

Berman has, as noted earlier, locked up an impressive string of Republican endorsements. The four highest LA County Republicans, LA County Supervisors Don Knabe and Mike Antonovich, as well as Sheriff Lee Baca, and Los Angeles County District Attorney Steve Cooley have all endorsed him. That's no easy feat, as I'm sure it can't be very comforting for them to know their name is just a little above Barbra Streisand.

The San Fernando Valley Republican Club is very influential. They meet every month at Galpin Motors. Jane & Bert Boeckmann, the owners of Galpin, also endorsed Berman. So it had to be disappointing that he wasn't able to run the table when the San Fernando Valley Republican Club chose not give an endorsement. I know several people who are with the club and I should find out what pitch Reps. Berman and Sherman gave. Whatever it was the club couldn't, in good conscience, endorse someone they agreed with on few issues.

Friday, August 17, 2012

Ryan Bounce?

Was there a Ryan bounce in the polls? It's tough to tell. Only two pollsters did national polls this week, Gallup and Rasmussen. Both have been kinder to Romney than most others. Gallup has been consistent, candidates within two, for over a month. Rasmussen went up and down. The week finished at Obama +1, better than he's been doing with Rasmussen.

There were two polls in Wisconsin, both with a Romney-Ryan bump. This should be expected and I think it's one state where it'll hold. People talk about how a VP doesn't bring his state, but that state is rarely one that swings. Republicans have shown the last two years that if they're motivated they can win Wisconsin.

There was only one Pennsylvania poll, from Franklin & Marshall. Romney-Ryan jumped 7 points from their previous poll, which is significant. They did have the biggest Obama margin previously. So we'll see if there's a bump from pollsters that had it closer.

Florida had two polls, from Rasmussen and Purple Strategies. These are the only two pollsters to show a Republican lead since May and they didn't show a Ryan bump in either of them.

Michigan has been an interesting state this year. Rasmussen, which usually has a Republican lean, has been one of Obama's best polls. The only poll out was from Mitchell Research a local pollster. And it showed a big Obama swing.

Ohio had three polls. Two of them were from pollsters who've leaned Romney, Purple and Rasmussen, and both showed bumps. The third was from PPP. They had a Romney jump in June and he remained steady.

Virginia's poll was from Purple and it had a nice Romney jump.

Purple's Colorado poll, however, moved two points towards Obama.

There were two polls in New Hampshire. One, from PPP, showed a big Romney jump and the other, from a local pollster, had a tiny one.

There was a Missouri poll from SurveyUSA. They hadn't polled the state before, but they showed the race as surprisingly close. This is even more surprising because Todd Akin, considered the weakest Republican candidate, had an enormous double digit lead over incumbent Claire McCaskill. People expect Romney to outpace Akin, not the other way around.

Conclusion: Rather mixed. There were some definite Romney bumps, but there were other polls that didn't show that. I don't think the bump, if it does exist, will last except in Wisconsin. The Republican National Convention is a week away and candidates often experience a post-convention bounce.

In 2008 Obama's bounce lasted until Palin's speech. McCain got a bounce that lasted until the fiscal meltdown. So Romney-Ryan may show some strong polls next week. Then the question will be how much each convention helps the candidates.

Tuesday, August 14, 2012

Dennis Cardoza has a new job!

I'm glad to hear it. I was worried he'd be out of work.

Rep. Dennis Cardoza resigns

Rep. Dennis Cardoza has resigned from congress effective immediately. His reasons are vague but we can call it short timer syndrome. There are only 7 legislative weeks left in the schedule and his party is in the minority. No one is counting on his vote.

Election Code 10703 is a little confusing. Based on the number of days needed, at least 112, a special election can't be scheduled before December 4. The governor is allowed to use his discretion and not call an election in the final year of the term. The law allows the election to be held if a regularly scheduled election is outside that range. I read this that it can occur 126 to 180 days to coincide with a general election, but not that it could be sooner. So the office may go unfilled.

Who would run in a special election? The Democrat running in Cardoza's seat in November is Rep. Jim Costa. Since Costa is already in congress, he can't run. The Democrats would have to find someone else. Perhaps perennial candidate Loraine Goodwin? I think she's up for another election campaign.

Friday, August 10, 2012

Obama-Romney polling

There have been several recent polls that have shown President Barack Obama way ahead. There has been significant criticism that these polls have too many Democrats and too few Republicans. That's certainly a good point, but there are a few polls that do drag the overall number down. Are the Bain attacks working? Is the polling moving toward Barack Obama? Most of these attacks are in the swing states. So, below are the swing states that have had one or more polls in June and July-August to see the change. Below that is the national polling broken down by month, separating July from August.

Of the 11 swing states which had polls both months we see that only one has moved towards President Obama by more than 2 points and 5 of the 11 haven't moved towards him at all. Likewise the national polling hasn't varied that much. I suppose Democrats could get optimistic that President Obama has been gaining every month, but adding 1.4 points over three months is within the margin of error and it certainly doesn't create a slam dunk win. If we don't go by month, and start on July 22, Obama's lead goes up 1 point to 4.9. That's not a knock-out blow by any means, although it's a healthy margin. Of course we got there by arbitrarily picking a start date, use several polls that critics believe are Democratic heavy, and ignore the state polls.

Wednesday, August 8, 2012

Pete Stark Remains Crazy

I'm not one to disparage people on a personal level but Pete Stark is crazy. Actually, I can think of a few words that better describe him that are more colorful but this is a blog I want you to bring your whole family to. And really what words could I come up with that'd be more descriptive than the ones that popped into your head when you read that article?

This district has a decent amount of Republicans and they shouldn't be a hard sell. They'd love to beat Stark. If he wraps them up, Swalwell needs 35-38% of the rest of the vote. I don't know much about Eric Swalwell but if he's reasonably competent he wins in November.

Sunday, August 5, 2012

State and National Polling: Is There a Disparity?

There's been some talk on the web about a disparity between national and state polling for the Presidential race. Mitt Romney trails Barack Obama by only 2-3 points nationally, yet he's only leading in two states Obama won in 2008, Indiana and North Carolina, and one of the 12 states considered to be swing states. One reason for this might be that Mitt Romney has only improved on John McCain by 4-5 points. There were only two additional states, Ohio and Florida, that Obama won by 4.5 points or less, Florida and Ohio. Barack Obama won Virginia, Pennsylvania, Iowa, Colorado, and New Hampshire by 6-11 points.

If all states swung uniformly, as posited in my previous post, Romney would not only need to win, but likely win by 2 points. Obama did that well in swing states. That aside, Mitt Romney is actually doing better than the 2.7 point national aggregate spread right now.

In fact, based on current polling the national margin should be around 2 points. Mitt Romney has improved in blue states by 6.4 points, in swing states by 5.1, and in red states by 4.7. Romney will need to do better in swing states in order win the electoral college. Flipping the other states that Obama leads by 3 points or less, Virginia, Florida, New Hampshire, and Colorado, he still won't result in a Mitt Romney win the election.

Unfortunately for Mitt Romney, many of the states he's doing the best over John McCain, Rhode Island, Massachusetts, Washington, Oregon, and Connecticut, are so blue that he won't flip them.

Saturday, August 4, 2012

Uniform Swing

In 2008, Barack Obama defeated John McCain by 53%-46%. If Mitt Romney beats Barack Obama by 1%, he'll have improved on John McCain by 8%. The uniform swing theory of elections is that the national popular vote swing would be constant across all states. In 2004, George W. Bush improved on his 2000 two-party performance by 1.5%. Thirty of the 51 states/DC swung within 2% of that 1.5%. They moved between 3.5% toward Bush and 0.5% toward Kerry. Forty of the 51 were within 3%.

The uniform swing tab in this spreadsheet has the states moving within 2% of the Bush-Kerry move in light blue and those within 3% in purple. In 2008, 26 states were within 2 points and 35 were within 3. Among the states that weren't were Massachusetts, Arizona, and Hawaii, states that were the home states for John Kerry, John McCain, and Barack Obama.

While using uniform swing is acceptable for projecting 2012, but we should keep in mind that 10-15 states will be more than 3 points outside of it. While Obama improved 4.7% over John Kerry, John McCain actually equalled or improved on George W. Bush in 5 states. If Mitt Romney gets more popular votes than Barack Obama and improves on John McCain by 7 points or more, he'll probably improve on 2008 in no more than 1-2 states and might not improve in any.

Wednesday, August 1, 2012

CA-9: McNerney Internal

Democratic congressman Jerry McNerney has put out an internal poll that shows him beating Republican nominee Ricky Gill in a landslide 49%-33%. Of course internal polls are suspect because the only reason to put one out is that you want the people with whom you're talking, to believe that you're doing better than popular opinion. Knowledgable people often have two reactions, adjust the poll by whatever you believe internal bias is or ignore the poll altogether. Even if you do these things, the lead is still in your head. Here are a few things to consider:

1. In 2008, the same pollster had McNerney winning by 10. He won by 1.

2. This was a D+4 district in 2008. That means Barack Obama did 4 points better than his national average. They have Obama leading 53%-38%. If it's 53%-38% Obama and remains D+4 then Obama is winning 49%-42% nationally. He isn't leading by 7 in any national poll. If it's D+4, then President Obama should be leading 51%-41%..

Wait a second, you might say. Maybe this district is moving to the left. Everybody says that California is moving to the left. 1. The Democratic registration advantage has fallen from 9.5% to 7.3% in the last year, the second biggest GOP gain in the state. There may be districts that are becoming bluer, but it'd be really odd in a district that Democratic registration has fallen by 1,700 people while Republican registration increased by 4,500.

2. Republicans won the primary 52%-48%. That'd mean McNerney has moved the district by 20 points and people are abandoning the Republicans in droves.

3. Republican primary turn-out was better than registration in the district, but not significantly as it was in a number of other districts. McNerney could certainly win in this district but winning by 16 would mean this district would go have a participation advantage well beyond their registration advantage.

Any of these things could be true, but there's no evidence to support it. Evidence supports that the opposite would happen.

Democratic Heavy Polls

Part of President Obama's lead in some polls is oversampling Democrats/Obama voters. Quinnipiac is out with polls today in Florida, Pennsylvania, and Ohio. They use a sample where Obama won Florida by 13 points, when he actually won by 3, Ohio by 15 points, when he won by 4, and Pennsylvania by 14, when he won there by 10. If you looked at these and the Party ID numbers, you'd think Democrats dominate each state.

Of course all three states have Republican governors and Republicans controlling both houses of the legislature. And this isn't a phenomenon brought on by the Republican wave of 2010. The GOP controlled most of those state legislative houses before then and did well in statewide elections in 2006.

If we use more realistic Party ID breakdowns, Obama's 6 point Florida lead becomes a 2 point Romney win. His 6 point Ohio lead is a tie. His 11 point Pennsylvania lead goes to 10, as Pennsylvania isn't as egregious.

It's the same thing for the senate races. They have Democrat Bill Nelson leading by 7 in Florida. A more realistic sample would have Nelson by 1. They have Democrat Sherrod Brown leading by 12 in Ohio. A more realistic sample would cut that to 5.