Monday, February 28, 2011

Who's Hurt By Redistricting? Further Analysis

California - There's been a great deal of speculating for California, with some of it focusing on Republican David Dreier and Republican Jerry Lewis considering retirement. A number of people have drawn maps in a non-partisan way and they favor Democrats. California is so big that you could draw compact districts ten times and come up with ten different maps.

Florida - Republicans are drawing this map, but Fair Districts Florida will hamper them. I'm pretty sure Republican Allen West is a goner because Palm Beach and Broward Counties are very Democratic. Any compact district is going to be very Democratic. Others will be in danger but speculation is also difficult.

Georgia - Republicans will try to eliminate Democrat Sanford Bishop by making his district more Republican.

Illinois - Holding 11 seats in a Democratic state where the Democrats are in charge of redistricting puts Republicans in a bad situation. Peter Roskum and Judy Biggert figure to be in the most trouble as their suburban districts border Democratic city districts. Mixing the city and suburbs should make their districts very Democratic. The further a Republican is from Chicago the safer he or she is. Outside of the East St. Louis seat it's difficult to draw a downstate Democratic seat. Bob Dold and Randy Hultgren are the next most likely to be in trouble, but Adam Kinzinger and Joe Walsh could see their districts become more Democratic.

Iowa - This could hurt all the Democrats but Democrat Tom Boswell will be in the most trouble.

Louisiana - Republicans Landry and Boustany figure to face off, with Landry more likely out.

Maryland - Andy Harris could be in trouble if the Democrats try to split the Eastern Shore into more than one district. Expect an ugly map.

Massachusetts - There's no indication of who might be in trouble yet. Since the delegation is all Democrats, it'll be a Democrat.

Michigan - I think Democrat Gary Peters is the most likely to lose his seat.

Missouri - Democrat Russ Carnahan shouldn't make long term reservations.

New Jersey - Speculation focuses on Democrat Rush Holt and Republican Leonard Lance, but New Jersey is less clear than other states.

New York - Democrat Carolyn McCarthy looks like a goner. Upstate you can speculate on almost anybody. The 26th is vacant now, but there will be a new rep there.

North Carolina - A Republican gerrymander could place Democrats McIntyre, Kissell, Shuler, and Miller in danger. That's not one of them. All of them could have more Republican districts with the same map.

Ohio - Democrats Betty Sutton and Dennis Kucinich are in the most trouble. A Republican might be hurt but it's too early to know whom.

Pennsylvania - Democrat Mark Critz looks like a goner.

Others will surely be hurt by redistricting, but these are the most likely.

Wednesday, February 23, 2011


Gallup is out with their latest poll of Obama approval ratings. Nate Silver at 538 is surprised at Obama's stronger than expected showing in Mississippi and Georgia.

If Mr. Silver had read this blog on January 14, he would've seen that Mississippi had the most dramatic improvement at the voting booth, compared to the average vote totals for the Democrats in each election. I flagged Georgia as a possible swing state. The high approval ratings in Maryland, Massachusetts, and Connecticut were seen at the polls. There isn't 100% correlation with the votes. The Democrats improvement in Arizona and Arkansas is not seen in the approval ratings, although his low 2008 Arizona total due to John McCain might have made the Democrats improvement look bigger than it. Likewise his popularity in Illinois doesn't translate to all Democrats.

Monday, February 21, 2011

Ronald Reagan and the 1994 Election

I’ve been perplexed for a long time at why the Republican wave in 1994 wasn’t followed by a decline in 1996. After 8 of the 12 wave elections the party winning the seat declined losing at least 20 seats in the following election. Twice the difference was in the high 20’s. Twice it didn’t occur in the following elections, but did so within three elections. These two times were during the Depression and Watergate.

Democrats got back only 9 seats in 1996 and it took them 7 elections before winning the seats back. These elections in 2006 and 2008, however, were their own wave and had a decline in 2010.

So why wasn’t there a decline? Ronald Reagan. Yes, Reagan. Reagan started a conservative revolution in 1980, taking in Christian Conservatives, “Reagan Democrats,” and an incredibly Republican youth. From a congressional standpoint it looked like it happened all at once in 1994. It didn’t. Republicans only lost 5 seats at the 1986 mid-term, the smallest mid-term loss since 1962.

There was no Reagan fatigue in 1988. As a result, George H.W. Bush won the Presidency and Republicans only lost 2 seats. At Bush’s mid-term Republicans lost only 8 seats. Bill Clinton had a big win in 1992. Such a win produces coat tails, yet Democrats lost 9 seats. This marked the 4th straight election producing results that are contrary to what happens almost every other time in elections.

Each time there was a Democratic pull the electorate was becoming more Republican, negating a big Republican loss. Then, in 1994, there were finally the conditions for a Republican election. The dam holding back what had been happening gradually burst and the Democrats lost 54 seats. By this point they weren’t Democratic seats that went GOP because of one election higher Republican turn-out. This was the new reality. Enough congressional seats were now GOP leaning for Republicans to have 225-230 seats.

Wednesday, February 16, 2011

The First Poll

A common theory is that if an incumbent is below 45% in a poll and that if he or she falls behind it's over. Here's a look:

The first group are incumbents who lost and the second group are those where the incumbent won. Clearly incumbents have come back from starting with less than 45% of the vote. and a number have come back from being down heavily. Harry Reid In the 2010 Nevada Senate race is the best example of that. Being behind does mean the candidate is more likely to lose, but that's what we'd expect.

Likewise, being way ahead is definitely an advantage, but enough lost after starting with double digit leads. Notable on this list are Russ Feingold and Blanche Lincoln. Their first polls, taken in the first half of 2009, looked good. It wasn't until much later in the cycle that their losses became apparent.

There may be a magic number that shows an incumbent is vulnerable. None who've had 52% or more in the first poll have lost, although Saxby Chambliss and Mitch McConnell got scares.

So there's no number that's the kiss of death and coming back from being behind is doable. The only correlation is that the better your numbers are, the better your chances are. But that one was obvious too.

Sunday, February 13, 2011

The Jungle Primary

California has adopted a jungle primary system for congress and the senate in 2012. Speculation is that this'll lead to more moderate candidates winning.

Georgia-9 is the most Republican district in the country. Last year they had a jungle primary in a special election. Six Republicans beat the Democrat 92%-6%. Tom Graves was elected in a run-off with fellow Republican Lee Hawkins. I believe Graves was the more conservative of the two. There'll be a lot of California districts that are heavily Democratic or heavily Republican. I don't think the centerist will win in them.

Hawaii-1 held a jungle special election last spring. The district is fairly Democratic, but not overwhelming like Georgia-9. The Democrats took 58% of the vote and the Republican 39%. The Republican and the more liberal Democrat finished 1-2 and then faced off against each other in November. Even in a Republican year the liberal won.

Florida had a similar general election this year for the senate, with a Republican running against a Democrat and a centerist former Republican. The election was equally split with Republicans and Democrats. Marco Rubio, the Republican, got 87% of the Republican vote, freezing out Charlie Crist from any moderate Republicans. Crist split Democrats with Meek, but was only able to take 36% of independents. Rubio took 51%. The more conservative candidate won.

In 2006 the moderate Democrat won a 3 way race in Democratic Connecticut and the moderate Republican won a 3 way race in conservative Alaska. In both cases the opposing party ran an uninspiring candidate. That enabled the moderate to suck up a lot of votes from the other party. There's no way of knowing how the candidates would do in a 2 way race with the entire electorate voting.

Washington had a top two primary last fall. All the incumbents but one finished first. Rick Larsen finished 0.2% back. They all faced the candidate who was expected to get the nomination from the other party in the general. They all won. The lone open seat, the third, drew five candidates who each got at least 11%. The party preferred Republican and Democrat finished 1-2 and faced off int he general. So there's no evidence that a moderate candidate will beat an incumbent or the conservative or liberal. In cases where the moderate won, e.g. the Rhode Island governor, he or she was always a high profile well funded candidate.

Thursday, February 10, 2011

Top Ten Senate Races

With the retirement announcements from John Kyl and Jim Webb, it's a good idea to look at the 2012 senate races.

10. New Jersey - Bob Menendez isn't a strong candidate. The Republicans, however, have yet to put anyone up against him.

9. Florida - Florida is trending Republican and the GOP looks to have strong candidates. Bill Nelson could be on shaky ground.

8. Nevada - John Ensign's affair damages him as a candidate. If Dean Heller is the nominee it'll move off the list. The Democrats have no one yet.

7. Michigan - Democrats got clobbered here in 2010. While there's no serious challenge to Debbie Stabenow yet, this race is worth watching.

6. Ohio - First race I'd definitely call a toss-up. Republicans have some strong candidates, but none have announced. Sherrod Brown is pretty far to the left and this state is trending Republican.

5. Missouri - Another state that Republicans should do well in 2012. The candidates so far are Sarah Steelman and Ed Martin. Both could put up a strong challenge.

4. Virginia - This is the first state I'd put as favored to flip. George Allen is in for the GOP, while the Democrats don't have a candidate yet.

3. Montana - I'm negative on Jon Tester's re-election chances. He won by the smallest margin in a Democratic year against a scandal tarred opponent. Danny Rehberg is a proven statewide candidate and this is a red state.

2. Nebraska - Cornhusker Kickback. Bruning. Need I say more?

1. North Dakota - Red state and Democrats have no bench.

Not on the list: Pennsylvania, Wisconsin, Massachusetts, Maine, Arizona, West Virginia - All these states except Arizona have strong incumbents. The GOP looks to have better candidates in Arizona.

Wednesday, February 9, 2011

The Myth of the Moderate District

Aaron Blake at the Washington Post talks about "The myth of the dying moderate." It's true that the moderate isn't dying but he's mistaken on this point:

What's more, when a moderate is kicked out of a swing district, he or she is often replaced by a moderate member of the other party -- or at least a member that learns quickly that he or she needs to be moderate if he or she wants to stay in office.

If you take a look at the membership of the Republican Main Street Partnership you'll see that the most common states for the congressmen are California, Illinois, Michigan, New Jersey, New York, Ohio, and Pennsylvania. Aside from California and Pennsylvania none of these states have blue dog members. Of the 10 new RMSP members only two replaced Blue Dogs. RMSP members come from moderate to liberal districts in moderate to liberal states. Only 4 of the 45 congressmen come from a state John McCain won. A lot of them replaced party line Democrats, who while not the most liberal in the Democratic caucus they weren't moderates either.

The Blue Dogs come largely from conservative southern, plains, or mountain states. It's heavily Republican country, many in states won by John McCain. So they were replaced by conservative Republicans. There are moderate districts. These districts usually only have a Republican moderate if it's a blue state like New York or Illinois and a Democratic moderate in a red state like Georgia or Louisiana.

The states with both are California and Pennsylvania, which are very blue and light blue. So it's not surprising that Republican moderates represent districts in them. Blue Dogs in Pennsylvania tend to represent rural Republican area. In California, however, you have a number of Blue Dogs representing liberal districts. I don't have an explanation for this.

Edit: I checked National Journal's 2009 conservative/liberal scores. If you click on Blue Dogs and Main Street Partnership you'll discover that the most liberal Blue Dogs and the most conservative Main Streeters are in California. These are moderate groups, so their voting records indicate they shouldn't be members. So, yes there are Blue Dogs in California but they're most liberal in the coalition.

Monday, February 7, 2011

Early House Prediction

Based on what usually happens when a President runs for re-election a good prediction would be a Democratic gain of 15 House seats in 2012 to cut the Republican advantage to 227-208. How Obama does is unlikely to change that. The losses will likely be less due to Republican gerrymandering.

Here's what I'm thinking now:

Benefitting the Republican Party
North Carolina –Right now Democrats have a 7-6 advantage in congressional seats even though this is a Republican leaning state. They have control of redistricting for the first time and it’s conceivable the GOP could draw a map where with 10 districts having a Republican lean. A more conservative map would give them 9 seats with a decided advantage. I’m going to be a little conservative and predict an 8-5 Republican delegation.
Republicans +2, Democrats -2

Georgia – Republicans can draw a pretty favorable map here too. It’s possible for them to draw out both Sanford Bishop and John Barrow, but that might be too ambitious. Instead, Republicans can draw a new Republican district and get rid of one of those two.
Republicans +2, Democrats -1

Utah – Republicans have tried to draw Jim Matheson out of a district and I think they’ll succeed this time. It shouldn’t be hard to draw four safe Republican districts.
Republicans +2, Democrats -1

Iowa – Iowa will lose a seat and that doesn’t bode well for the Democrats. There are two strongly Republican districts and three weak Democratic ones. When you make these five districts into four, you could end up with four Republican leaning districts. I think the GOP will take either the 1st or 2nd from the Democrats and win the 3rd with Latham running against Boswell.
Republicans +1, Democrats -2

West Virginia – This state keeps getting worse for Democrats. Nick Rahall survived this year due to weak competition. The state will go heavily Republican in the Presidential race and that should be enough to doom Rahall.
Republicans +1, Democrats 0

New York – While Republicans and Democrats both will have a say in redistricting, I think it won’t bode well for the Democrats. The state loses two seats. Republicans only control two New York area seats, so the Democrats will be squeezed out of one. Upstate there’ll likely be several toss-up seats, with one pitting a Democrat against a Republican. I’m giving the GOP an edge there.
Republicans 0, Democrats -2

Massachusetts – Democrats have all ten seats and the state is losing one. The math only works one way.
Republicans 0, Democrats -1

Michigan – Republicans control redistricting and that should be enough for them to gerrymander a Democrat out of his seat.
Republicans 0, Democrats -1

Missouri – Popular speculation has Republicans and Black Democrats teaming up to save both Black Democrats and sacrifice Russ Carnahan.
Republicans 0, Democrats -1

Kentucky, Connecticut, Oregon, and Oklahoma all present opportunities for the GOP. Republicans will probably pick up at least one seat in these states.
Republicans +1, Democrats 0

Benefitting the Democratic Party
Florida – Republicans hold a 19-6 edge in the congressional delegation, a ridiculously enormous edge in a state where their share of the statewide vote would result in something like 14-11. The good news for the GOP is the state is adding two districts and the Republicans control redistricting. The bad news is that Fair Districts Florida won’t allow them to gerrymander the way they like. Right now there are three districts in Palm and Broward Counties that are D+1, D+13, and D+15. Without gerrymandering these will all be Democratic districts. Maybe I’m a little pessimistic for the GOP, but a 17-10 edge would still be pretty significant.
Republicans -2, Democrats +4

Illinois – Republicans hold an 11-8 edge in a state that is heavily Democratic. Democrats should be able to squeeze four downstate Republican districts into three and then make three of the seven Chicago area seats Democratic enough to take them.
Republicans -4, Democrats +3

Texas – Like Florida, Republicans hold a disproportionately high edge in congressional seats. Here it’s 23-9. To make those 23 seats safe, Republicans will have to pack Democrats into new districts and sacrifice three of the four new seats. If they end up with a 24-12 edge, they’ll have done well.
Republicans +1, Democrats +3

California – I have no idea how the redistricting commission will impact the delegation. I wouldn’t be surprised if the Democrats picked up five seats or the Republicans picked up three. There are strong arguments for both. I’ll choose something in between for now.
Republicans -2, Democrats +2

New Hampshire – Both districts are toss-ups, with the GOP controlling both right now. Odds are that they’ll lose one or the other.
Republicans -1, Democrats +1

Virginia – With Democrats still in charge of one House in the Virginia legislature the GOP will have a tough time creating districts where they’ll hold onto all their gains. It’s an 8-3 delegation now. A 7-4 edge will still be significant.
Republicans -1, Democrats +1

Arizona – The new seat could go either way, but I think there’ll be a new Hispanic majority district that will go Democratic.
Republicans 0, Democrats +1

Nevada – Republicans hold a 2-1 edge despite this state trending Democratic. If they can make Joe Heck safer, they’ll let the Democrats create a Democratic leaning district.
Republicans 0, Democrats +1

South Carolina – While it’s possible to create six Republican districts, I think the Justice Department will insist on them drawing a second majority minority district.
Republicans 0, Democrats +1

Washington – Population growth has been heavily in Democratic Seattle. A new district there will likely make the 3rd and 8th more Republican.
Republicans 0, Democrats +1

Ohio – Republicans control redistricting but currently have every district outside of Northern Ohio. There are too many marginal districts. Republicans may need to draw a Democratic district in Columbus, while only eliminating one Northern Ohio Democratic district. Being conservative will help Republicans retain the seats throughout the decade.
Republicans -2, Democrats 0

Louisiana – The state will lose a seat. Since it’ll retain the majority-minority district in New Orleans, it’ll be a GOP congressman who loses out.
Republicans -1, Democrats 0

New Jersey – Another one that could go either way. I have a feeling that the Democrats will come out on top.
Republicans -1, Democrats 0

Pennsylvania – Republicans control redistricting here too, but also have a high number of marginal seats. While they’ll likely squeeze the Democrats out of a seat in Western Pennsylvania, they might not be able to gerrymander that they won’t lose an Eastern Pennsylvania district.
Republicans -1, Democrats 0

113th congress: Republicans 237 Democrats 198 (Democrats +5)
I could see how Republicans could get as many as 262 seats or as few as 215, but I consider neither very likely.

Left Wing Media Bias

Conservatives often complain about left wing media bias and the left disputes this. Here is an interesting informative article, that appeared yesterday. It is accurate and some might not be able to detect bias. Subtle bias is bias.

1. The article implies that Republican congressmen only take the Koch advocated position because of money from the Koch brothers. If Republicans who haven’t taken money from the Koches largely had different positions, that’d be good journalism. I’m pretty sure that almost all Republicans hold this position regardless of contributions and have held it long before they got any contributions.

2. The article implies that the Democratic position is the best one for the American public and that the Republican position is only good for the oil and gas industry. The environment is a complicated issue that should be open to discussion whether regulations go too far and what the impact of the regulations will be. Stiffer regulations means business spends more money. They either raise prices or lay off people to compensate.

3. The article implies that the Koch brothers are responsible for the Democratic majority, Nancy Pelosi’s position. How many Republicans did they actually give significant money to and what was the percentage of the overall money spent on those races? How did the Democrats in those races spend? How does their spending compare to other outside spending, like unions, and spending by the Republican and Democratic parties? Republicans took the majority with 24 excess seats. I don’t think the Koches spent money in that many winning races.

4. The article implies that the Koches are behind the Cato Institute, the Federalist society, and the tea parties, without indicating what percentage of money spent comes from the Koches. The tea parties existed before the Koches got involved and their involvement is limited to Freedomworks providing assistance to tea parties. Tea parties don’t accept a heavy hand from anyone.

5. The article implies something sinister about the Koch conference without providing evidence or noting that there are thousands of such conferences put on by businesses every year.

6. The article waits until the last paragraph to note that the Democratic members of the committee are liberal environmentalists. Is this true? Did Henry Waxman replace those that weren’t? Environmental groups employee lobbyists, just like business groups do. They spend money, just like business groups do. Do business groups spend more money? No idea. Wouldn’t that be a nice thing to investigate?

Bias doesn’t just come from advocating a point of view. It comes from how you present the facts, what you support, and what you choose not to support. Journalism doesn’t have to be balanced, but it needs to be thorough and put in context. The Koches may be just as sinister as they make them out to be, but by leaving out important information all they’ve done is write an article which will fuel liberal paranoia and anger conservatives.

Sunday, February 6, 2011

Did the Tea Party Hurt the Republicans in the Senate?

Conventional wisdom is that the tea party candidates cost the GOP at least 3 senate seats because they were weaker. There’s no way to know what would’ve happened in Colorado or Nevada, but Delaware would’ve likely flipped. Mike Castle was an unusual candidate. How would Christine O’Donnell have compared to anyone else?

I decided to compare the states that had similar 2008 Presidential spreads.

With Joe Miller we know that Lisa Murkowski was the better candidate since she beat him and the Democratic candidate. Mike Lee didn’t exceed John McCain’s 29 point 2008 margin, but exceeding that wouldn’t have been easy. John Boozman and David Vitter didn’t do any better. I doubt Bob Bennett would’ve been a better candidate. Of course they would’ve won this regardless.

Kentucky, Kansas, and West Virginia all had similar 2008 spreads. Jerry Moran won in a landslide, while Rand Paul won by a smaller margin and John Raese lost. There’s hardly enough data here to make a judgment, especially since Paul and Raese had serious challengers, and West Virginia is a far tougher state for a senate race than a Presidential one.

The last group doesn’t have a tea party candidate. They all won in lopsided races.

Marco Rubio is considered a tea party candidate, although his background as Florida Speaker of the House hardly makes him an outsider. He had a larger margin of victory than any other Republican in a similar state.

The next six states were somewhat more one sided for the Democrats in 2008. Charles Grassley was an incumbent, so he’s not a good comparison. Kelly Ayotte was an establishment candidate and her spread was next best. Ron Johnson and Pat Toomey, both considered tea party candidates, did very well. Ken Buck and Sharron Angle did a lot better than McCain, but were the worst of the group.

The last group were Obama landslide states. Tea Party favorite Christine O’Donnell exceeded McCain by 8 points, a total better than many of these candidates. Mark Kirk, an establishment moderate, fared best, while Dino Rossi did well too. Most of the rest couldn’t be considered establishment or tea party, although Carly Fiorina and Linda McMahon got the nod over a tea party candidate and a more establishment one.

It’s impossible to say definitively how the tea party candidates did, since the circumstances of individual races was a big factor. In most cases the establishment candidate raised more money and in some had a weak challenger. For the most part, however, the tea party candidates fared worse than the establishment candidates. So there’s likely some credence to the argument the tea party candidates hurt the GOP.

Thursday, February 3, 2011

Senate Coat Tails

Public opinion is that Obama helped Democratic performance. It's tough to argue with that. Obama had stronger showings than Kerry in most states and the Democrats won 20 of 35 Senate contests. How much help did Obama provide and more importantly, was it important for Obama to win a state to push the Democrat over the top?

Democrats did better in 21 of the 28 contests where both elections had a Republican and Democrat. Of course some elections had better Democratic candidates or better Republican candidates. Democratic recruitment was better in 2008 and Republican recruitment was worse, so we'd expect the chart to skew toward 2008. It's possible that the environment would've helped the Democrats, regardless of who ran for President. The overwhelming number of elections where the Democrats did better was so great that Obama's coat tails played some role, if not the entire role.

Let's look at Obama's percentage of the vote compared to the Democratic candidate to see if there's a correlation.

I'm no statistician, but the number of elections where Obama and the Senate candidate were close indicates some correlation. The races in red were open seats. In most of these the Democrat adhered closer to Obama's percentage. The races in yellow are those where the candidate lost the state but the party's candidate won. Here, the correlation isn't that high. It's likely that Obama helped the candidates, since the Democratic candidate won all of these states. There's something even curiouser, however. West Virginia and Louisiana were 2 of the 3 states where Obama did worse than Kerry. Obama wasn't a drag on either of these races. The Democratic incumbents won anyway. So Obama won't necessarily be a drag on Joe Manchin or Jon Tester in 2012.

Let's look at the last President's re-election campaign. Republican candidates heavily underperformed George W. Bush in 2004.

Bush may not have pulled his party's candidates to victory. Obama was a positive in most Senatorial races in 2008. That doesn't mean he will definitely be one in 2012.

Wednesday, February 2, 2011

Senate News

In the last week or so we've had Kent Conrad's retirement, Joe Lieberman's retirement, Denny Rehberg entering the Montana senate race, and a poll showing Jon Bruning is ahead of Ben Nelson by 11 points.

I'm upping my prediction of Republican gains from 5.1 to 5.4 seats. That's not a lot but I already had these three seats as the most likely to flip. Since the GOP only needs 4 seats for the majority, there's probably about a 65-70% chance the chamber flips.

Tuesday, February 1, 2011

Getting it

Charlie Cook, an elections expert, kind of gets it. He understands what I've been saying for the last two months. Democrats regaining the House is highly unlikely, although not impossible. The circumstances need to be right for a wave election and it's doubtful the climate will change into those circumstances. He mentions a government shut down, but that happened in 1995 and the Republicans only lose 8 seats in 1996. Wave elections occur almost always as a negative reaction to the sitting President or with an election of a new President. The Democrats can't benefit from either of these scenarios.

That's not to say the Democrats won't pick up seats. I'm pretty sure they will. I'd say they'd end up around 207 after the 2012 election if Obama is elected and 195 if he loses.