Friday, April 25, 2008

West Michigan State House Races

(I cross-posted this at West Michigan Rising earlier this week to celebrate the end of exams. If you are a Democrat from west Michigan, check out the site-PB)


Thanks to my exams being finished at Michigan, I am doing a bit more of the number crunching I promised in the past two months. I am working with Mark Gerbner and Michigan Liberal to start doing some detailed analysis of the Michigan House races. I hope to provide some detailed analysis of the swing seats (16 or so) for Michigan Liberal, but this analysis probably won't finish up until later in May after the filling deadline closes. However, I did want to provide some early analysis on west Michigan races to promote some discussion on West Michigan Rising.

West Michigan in my mind goes from Traverse City in the north to Sturgis in the south and includes everything to the west. With these parameters, there are 27 Michigan House seats in the region. Since we have a number of strong candidates and incumbent representatives starting to launch campaigns as the snow finally melts, I wanted to see what baseline we might be operating with in this election cycle.


My basic unit of measurement is based on Charlie Cook's Partisan Voting Index (PVI). Cook uses PVI to find how strongly United States Congress districts lean to one party over the other. Cook uses the district's presidential vote totals from the past two elections and finds the average. The average is then subtracted from the national popular vote average. For example-Vern Ehlers district has a R +9, meaning that the 3rd congressional district is 9% more Republican the rest of the nation.

For this analysis I modified Cook's formula slightly. I used Mark Gerbner's Democratic baseline vote, which he defines as the average Democratic share of the vote cast for major parties for the four statewide education boards. Mark helpfully broke these totals by Michigan House and Senate Districts (I'll analyze Senate Districts at a later time). I then took the 2006, 2004, and 2002 baselines, and found the average, which I subtracted from the statewide baselines. From this I found the PVI.

Before discussing the PVI for these districts further, I should note that as Mark mentioned in Michigan Liberal postings, Michigan has become an increasingly Democratic state since the 1998 elections. This is especially true since 2002, as the Democratic baseline increased from 49.2% in 2002, to 52% in 2004, and jumped to 55% in 2006. Since the PVI is based on this statewide average, districts that have small Republican PVI are more competitive than they first appear, especially since to win one only needs 50%+1 vote.

the full spreadsheet is here
Table One shows the data for west Michigan's 27 House seats and are ranked from highest to lowest Democratic PVI. While the Democratic baselines have increased by at least 4% in most districts, some sport a double digit gain in their baselines (75th). Interestingly, the districts that are the most marginally Republican (Swing, Weak, or Lean Republican) are the ones that have had the greatest increase in their Democratic baselines. This trend is especially noticeable in some districts where we have top-notch challengers.

Table One also shows what districts are the one most open to a Democratic take over in 2008. Conversely, we can also assume which districts our party should devote some serious efforts to define (of course, we all know which ones these are already, but it doesn't hurt to think about it again). Below I have grouped the districts in different status groups, and I have included some maps to provide some further reference for interested readers.

Safe Democratic (1) District 92-This Muskegon-based district has never sunk below a 62% baseline. Nothing to worry about here.
larger southern districts map here
Strong Democratic (1) District 60-Robert Jones Kalamazoo seat should not face a serious challenge with a PVI of D+9.

Leans Democratic (1) District 76-This west-side Grand Rapids district has long been Democratic, although I can remember some fierce races back in my youth (I'm thinking of the 1996 race between Pestka and Kozak). Since Sak is term-limited here, this would be the best chance for the GOP to pick up a seat. However, Roy Schmidt seems to be well-established to keep the seat in the Democratic column, as the Republicans have no standard bearer yet.

Weak Democratic (1) District 62-This Battle Creek-based seat is perhaps the best pickup for the Democrats in west Michigan. Term-limited Mike Nofs is leaving a seat that has had a majority Democratic baseline in the past two elections, and which Granholm and Kerry carried by even larger margins. This is a district that should receive serious money and a strong candidate.
larger northern districts map here

Swing (4) Districts 71, 75, 80, and 91. Democrats control the 75th and 91st districts, and from the early buzz, appear to be well-positioned to keep these seats. Both Dean and Valentine have ample funds, and the former faces especially marginal opposition. Valentine will face Holly Hughes, who has pushed a lot of personal money in the race. Whether this can result in a serious efforts remains to be seen.

The two Republican swing seats are interesting. The 71st District (covering Eaton County) is represented by Rick Jones, who is in his second term, and has a PVI of R+5. Still, the Democratic baseline has jumped in this district from 45% in 2002 to 51% in 2006. A good candidate might be able take this seat. The 80th District, covering Van Buren County and a few townships in Allegan County, has a PVI of R+6. Granholm carried this district twice, and a serious effort by Jessie Olson could make this seat very competitive.

Weak Republican (3) Districts 61, 63, and 101. These three seats all have a PVI of R+7 and are all open seats, making them very marginally Republican seats. Districts 61 and 63 (both based in Kalamazoo County) have serious Democratic candidates running, as does District 101 (Mason, Mainstee, Benzie, and Leelanau Counties). From the larger trends, it appears that 2008 is a Democratic year on par with 2006, so our candidates in these races are in the position to tip these seats into the Democratic column.

Leans Republican (3) Districts 70, 78, 100. These seats are more strongly Republican, each with a PVI of R+9. District 70 (which covers part of Ionia and all of Montcalm Counties) and District 78 (covering Berrien and Cass Counties) are open seats that have competitive Republican primaries and ready Democratic candidates. District 100 will likely be a greater challenge, since we don't have a Democratic candidate to run against a Republican incumbent. These seats are likely to remain in the Republican column, although a strong Democratic effort could do provide some serious double takes.

larger central districts map here

Strong Republican (5) Districts 59, 79, 87, 102, and 104. These seats located around Traverse City (104), Cadillac (102), St. Joseph (79), Cass and St. Joseph Counties (59), and Barry and Ionia Counties (87) are strong Republican seats. The good news is that we are running Democratic candidates in these races, something that does not always happen.

Safe Republican (8) Districts 72, 73, 74, 77, 86, 88, 89, and 90. These seats are the Republican base in Michigan politics. Covering Allegan, Ottawa, and the suburbs of Grand Rapids, these seats are places were lackluster Republican presidential and gubernatorial candidates come to hang out and stir up votes in hopes to getting a vote margin to hide under when the Wayne County returns arrive. Republican candidates in these seats often get free passes, allowing the GOP to direct resources and party infrastructure in these areas into critical districts. Keeping resources tied down in these districts is something that we desperately need to do to ensure that our party can offer strong challenges in other districts. That we have two competitive candidates running in the 73 and 88 Districts is a good start to ensure our Democratic majority in the Michigan State House.


As we begin the primary season, Democrats have some opportunities to expand the traditional playing field in west Michigan. While defending our swing seats in the 75 and 91 Districts should be a priority, supporting competitive challenges in Republican-held swing seats and weak Republican districts should be strongly considered.

I would love to hear feedback on this analysis-either on West Michigan Rising or via my email. I have all the precinct data for these years as well, so if anyone want me to crunch things further, please do not hesitate to ask.


Cory Smidt said...

A couple initial comments Peter.

1) '06 is a highwater mark for Democratic voting margins. Republican turnout was low and those trends you report are true for most every state. Specific to MI - the tanking economy also helped the local Dems in '02 sort of negating the Rep's strong national performance. This is going to bias your measure that includes '06, especially for an Education board vote split. Take district 71 - consistent 54% Rep in 04 and 02, then in 06 it dumps below 50%. I'd be surprised if that number doesn't go back over 50.

2) If you want to get at those places truly trending Dem, what you are going to need to do is distinguish true Democratic trends from the specific bump of 2006. A simple way may be to estimate the average bump in '06 (either across all districts or by district category) and then subtract (or add) that number from the Dems (Reps). Plus, you could look at what districts show the greatest variability from year to year. Sometimes it's not so much the level but the tendency of districts to swing wildly that are good for gains.

3) Not completely related, but you may find this interesting. There's an emerging literature in the Political Science State Politics field on whether term limits inhibit or enhance partisan margins and normal voting. This one is an example.

PB said...

Hey Cory,

Thanks for the comments-I'm glad that you are reading these articles.

1) I wanted to keep a balance with the elections. Yes, 2006 was a Dem year, just as 2002 was a GOP year, and 2004 was (on the state level). Personally, it seems that the political climate in 2008 is much closer to the climate in 2006 for the GOP, and perhaps even whose. I disagree with your comment that Republicans did worse the Democrats in 2002-they kept all their state house seats and won two very competitative open state senate seats that were toss up. In any case, the baseline vote is much lower than the Levin/Granholm vote (two Dems that did very well).

2) Very true. I think that I looked at the variablity to some extent, but perhaps I need to take a close look.

3) Interesting. I wonder what impact this might have on a year that 44 state reps are leaving (and about 30% from very close seats)

I hope that the move to Lansing goes well!