2020 Census and Redistricting Thread: Michigan

[lfromnj]

https://twitter.com/Redistrict/status/1457535235139244034?t=Oj2vnnHLBpSMw08fAZ1e...

I'm surprised the Grand Rapids-Kalamazoo district made it.  The other two maps are pretty normal really.  

They made Kildee's seat Biden+2 and the Lansing district close to 50/50 in all three maps.

The Birch map is probably the best for Democrats long-term as Grand Rapids and Upton's district are both trending D overall, so potentially up to 9 districts total.  For 2022 the best map for D's is the Grand Rapids-Kalamazoo "Apple" map.

I would have to imagine the MIGOP is rooting for the Chestnut map.

TBH, I feel like Meijer is someone who could become entrenched and be a consistent overperformer a la Fitzpatrick or Katko.

Pretty sure he underran James.

[lfromnj]

Yeah compared to legislative maps where some pretty absurd districts had to be done to make it "fair", nothing as absurd needed to happen congressionally but they still tried for the 0 efficiency gap. Muskegon with GR or Kal with GR is fairly absurd while Lower Macomb with Ferndale is bad but not nearly as bad as the other 2 so Birch is probably the best from a COI perspective.

[Torie]

I watched the MI redistricting commission in action on youtube today (a meeting last Thursday). First, the initiative that it operates under is poorly drafted, and the commission had 3 lawyers arguing, none of whom got it right. I think later on after I fled, somebody decided to hire a high powered law firm to tell them what I could have told them for free but whatever (because the meeting the next day mentioned that a nationally famous law firm had been hired). No, the commission does not have the power to interpret the law, and yes, the constitution may force changes in their deadlines and schedules, and maybe the date of the primary election, and yes anyone can sue with standing to sue, like an aggrieved commissioner, and yes there may be a necessity to file a declaratory relief action, and yes, as a practical matter, who is going to vote for an un-vetted map, and yes it is a mess.

Second, they have partisan fairness software that can look at election by election. What that means is that the geographic Dem disadvantage in Michigan, and the VRA, means that the map needs to be gerrymandered so that the Dems can catch up. Thus uber Dem Ann Arbor is quad chopped on the state house map, to spread the Dem wealth. The hapless Pubs have no idea what hit them. And finally, lawyers always take over. The chairperson happens to be a lawyer, claims she is unaffiliated but she is really a Dem (Michigan has no partisan registration), and she dominates everything.

I wonder if I will live long enough to see a commission go into the fairness metrics in states where geography favors the Dems, and the gerrymandering needs to be ruthless to get the Pubs to catch up. Such a state house map in Mass would be an erose mess.

The MI constitution does have a clause about using standard measures of fairness. Just what are standard measures is in the eye of the beholder I guess (and the mathematics is way above what 99.9% of the population can understand, so that just leaves Muon2 on this site), and such standards can rapidly lead to the theater of the absurd depending on what is going on on the ground. I won't be watching this circus again. It's just too painful.

Oh, on the CD level, the fairness measures are not too bad in MI vis a vis ugly gerrymandered maps, given that so much of MI has become much more marginal than heretofore, as Dutch Michigan and high income educated burbs race to the Dems, and rust belt whites race to the Pubs, rendering vast swaths of MI marginal. Most states are not like Michigan.

Oh, one other thing. If they go to ranked choice voting if they don't get a bipartisan agreement, the law is unclear if that vote is by secret ballot or how that will work, and if by a public roll call, giving those who vote last and have a high powered computer that they know how to use, a huge advantage, and whether one can change one's vote after knowing what the totals are, and if one can legally bargain behind closed doors. The commission has not though about that at all, although the smart lawyer chairman probably has. But she shows her hand only when it maximizes her power.

tldr: It's cf city out there in the land between the lakes.

Addendum 1: I see the post above that I just read is on the same wave length as I on much of this. So we are both sage and perspicacious, or clueless and obtuse.

Addendum 2: We should adopt the German system which mixes districts and proportionality, so it does not matter much how the lines are drawn. If the lines zero out a party that got 40% of the vote, they then get enough seats through proportionality to get to 40% of the seats. Yes, to make it work reasonably, we need to get rid of the Senate and the States. Both have to go. It's time for a Constitutional Convention to make it happen. I nominate myself as chairman of that event. It will need a lot of competent security guards I understand. Thus the Capitol police will not be used.

Addendum 3: Since the fairness doctrine in NYS requires that the Pubs get 40% of the CD's, and everything else. I wonder what that map would look like? Probably need to take the Pub map that was submitted, and do some more gerrymandering on Long Island, and do a snake from Rochester to Ithaca.

[wbrocks67]

7-6 Biden seems doable, and better than the current lines. Seems crisis averted here? I'm a bit surprised we didn't get an 8-5 Trump map or something.

[Pericles]

This looks beautiful, elections should reflect the will of the people so it's nice to see partisan fairness getting high priority.

[Antonio the Sixth]

7-6 Biden seems doable, and better than the current lines. Seems crisis averted here? I'm a bit surprised we didn't get an 8-5 Trump map or something.

Partisan fairness was explicitly the #1 metric the state constitution imposed on the commission, so this is not surprising at all.

Anyway, it also seems they did a decent job representing COIs on top of it, so that's a relief. Like the PA map last cycle, those are all largely fair maps which make small choices in favor of Dems when possible in order to correct for the Republican geographic bias. I think that's the right approach. Of the three, I guess I agree with ilfromnj that the "Birch" plan seems slightly better, though all 3 maps would be acceptable.

[Torie]

FWIW, I like Birch the best as well. As I say, given the fairly flat partisan variations outside Detroit and Ann Arbor, for a map with districts as large as CD's, it does not upset the apple cart that much. It does at the state house seat level however with much smaller districts that are easier to gerrymander.

[Torie]

Here is the C map I would have pushed I think if I was on the commission. Not much different from Chestnut, except the Grand Rapids CD is swing, rather than a ugly Dem snatch gerrymander, but Grand Rapids is going Dem anyway, so this iteration would have given the Pubs a short term lease, unless the Pub incumbent there was MAGA'ed out. The other changes from a partisan perspective are minor. I just don't do chops to help the Dems at the margin a point or less.

[ProgressiveModerate]

Honestly I think Birch is the best. It achieves partisan fairness without doing anything weird in Western MI with Grand Rapids to make a Dem leaning district.

Overall I'm happy with the MIRC. At first as was hesitant because it seemed like they had no clue what they were doing but in the end they came up with 3 really viable maps without too much tension between the Ds and Rs in the process.

[President Punxsutawney Phil]

https://davesredistricting.org/join/14274625-6937-4b41-a5b6-0a290d161505
MI state senate map with a 0.00% efficiency gap, but good county integrity. If need be, Washtenaw can be unified, with the Adrian SD taking in Jackson instead of southern Ann Arbor.

[Antonio the Sixth]

When is the Commission going to decide on a final map?

[ProgressiveModerate]

When is the Commission going to decide on a final map?

IIRC dec 21 but don’t quote me

[lfromnj]

https://mobile.twitter.com/LaurenMGibbons/status/1473013466176839681

[lfromnj]

https://mobile.twitter.com/Dave_Boucher1/status/1473327888426577924

[_.]

I tried my hand at a fair congressional map of Michigan using the 2020 census results.

Image Link

The Population Deviation is less than 0.01%.

52/100 on Dave's Proportionality Index
75/100 on the Compactness Index
26/100 on County Splitting
67/100 on the Minority Representation index
26/100 on Dave's competitiveness index

The map above shows results from the 2020 U.S. Presidential election in Michigan.

Check it out here and see county and municipality boundaries.


Partisan Breakdown by Election

2016 U.S. Presidential Election in Michigan: 8R to 5D

2018 Michigan Attorney General Election: 8R to 5D

2018 U.S. Senate Election in Michigan: 8R to 5D

2018 Michigan Gubernatorial Election: 8R to 5D

2020 U.S. Senate Election in Michigan: 8R to 5D

2020 U.S. Presidential Election in Michigan: 8R to 5D


One majority-Black district (50.8% VAP) and one that's 42.7% Black by VAP.

Every single election returns 8R - 5D but with drastically different margins. In the 2020 election, the closest district was the Grand Rapids one at Trump+0.4% (down from Trump+9.8% four years prior). The Central Oakland District is also interesting as it swung from Clinton+6.8% to Biden+13.6%. Notably, Asians (14.1% Population and 12.9% VAP) are the largest minority in that district and not Blacks.


Opinions?

[Antonio the Sixth]

When is the Commission going to decide on a final map?

IIRC dec 21 but don’t quote me

Well, it's Dec 26th and I am quoting you

[BoiseBoy]

The commission will be meeting a few times between December 28 and December 30.

We'll see a final map out of it. My favorite is the Chestnut plan.

https://twitter.com/RedistrictingMI/status/1475483860913864708?s=20

[The Undefeatable Debbie Stabenow]

Of course, I still like my map (https://districtr.org/plan/96349) more than any of these. That said, these final options are far superior to some of the absolutely hideous options that the commission was pondering earlier on. (It was also quite predictable that they would rather go for 2 majority-minority districts than take my approach, even though mine should still be totally VRA-compliant, so I'm not surprised about that.) I wouldn't be too upset with any of them, even though the Kalamazoo-GR district in the Apple plan really is quite dubious and Wahlberg's district reaching both Lake Michigan and Lake Erie is not really ideal as far as compactness goes.

[President Punxsutawney Phil]

Of course, I still like my map (https://districtr.org/plan/96349) more than any of these. That said, these final options are far superior to some of the absolutely hideous options that the commission was pondering earlier on. (It was also quite predictable that they would rather go for 2 majority-minority districts than take my approach, even though mine should still be totally VRA-compliant, so I'm not surprised about that.) I wouldn't be too upset with any of them, even though the Kalamazoo-GR district in the Apple plan really is quite dubious and Wahlberg's district reaching both Lake Michigan and Lake Erie is not really ideal as far as compactness goes.

https://davesredistricting.org/join/75dcd34d-39e2-4ff4-b049-bcc9d3284f42
thoughts on this map?
It's inspired by your map (which solves some problems I have had difficulty with), while also avoiding a whole-Detroit CD. It takes a less dogmatic approach to county splits than I usually do.
It also keeps Battle Creek and Kalamazoo together (not sure how much of a positive that is).

[bagelman]

Of course, I still like my map (https://districtr.org/plan/96349) more than any of these. That said, these final options are far superior to some of the absolutely hideous options that the commission was pondering earlier on. (It was also quite predictable that they would rather go for 2 majority-minority districts than take my approach, even though mine should still be totally VRA-compliant, so I'm not surprised about that.) I wouldn't be too upset with any of them, even though the Kalamazoo-GR district in the Apple plan really is quite dubious and Wahlberg's district reaching both Lake Michigan and Lake Erie is not really ideal as far as compactness goes.

You appear to have an 80% black district which is packing.

[The Undefeatable Debbie Stabenow]

Of course, I still like my map (https://districtr.org/plan/96349) more than any of these. That said, these final options are far superior to some of the absolutely hideous options that the commission was pondering earlier on. (It was also quite predictable that they would rather go for 2 majority-minority districts than take my approach, even though mine should still be totally VRA-compliant, so I'm not surprised about that.) I wouldn't be too upset with any of them, even though the Kalamazoo-GR district in the Apple plan really is quite dubious and Wahlberg's district reaching both Lake Michigan and Lake Erie is not really ideal as far as compactness goes.

You appear to have an 80% black district which is packing.

Are you suggesting that it's VRA-noncompliant? Someone can correct me if I'm wrong, as I'm far from an expert and the VRA is certainly very complicated, but I do not think that my map would violate the VRA's redistricting requirements. Yes, it's possible to create 2 majority-minority districts, and I only created one. However, the surrounding districts with significant minority populations (grey and maroon) will consistently vote to elect candidates that are preferred by minority voters, and consequently I don't believe there are any issues with the VRA. Racially polarizing voting is the critera here; the VRA's requirements are not nearly as simple as "if you can possibly draw X majority-minority districts, you have to draw X majority-minority districts." While this is certainly a great concentration of minority voters in one district, I don't think you could argue that this map dilutes the political preference of minority voters.

Now, as I stated in my original post, I fully expected the commission to default to drawing 2 majority-minority districts. It's a more straightforward way of ensuring VRA compliance, and a more straightfoward way of indicating to the general public that minority voters are securing adequate representation.

[The Undefeatable Debbie Stabenow]

Of course, I still like my map (https://districtr.org/plan/96349) more than any of these. That said, these final options are far superior to some of the absolutely hideous options that the commission was pondering earlier on. (It was also quite predictable that they would rather go for 2 majority-minority districts than take my approach, even though mine should still be totally VRA-compliant, so I'm not surprised about that.) I wouldn't be too upset with any of them, even though the Kalamazoo-GR district in the Apple plan really is quite dubious and Wahlberg's district reaching both Lake Michigan and Lake Erie is not really ideal as far as compactness goes.

https://davesredistricting.org/join/75dcd34d-39e2-4ff4-b049-bcc9d3284f42
thoughts on this map?
It's inspired by your map (which solves some problems I have had difficulty with), while also avoiding a whole-Detroit CD. It takes a less dogmatic approach to county splits than I usually do.
It also keeps Battle Creek and Kalamazoo together (not sure how much of a positive that is).

Looks really good! I had personally gone into the drawing process with the idea of keeping every single municipality, including the city of Detroit, undivided, which was not at all a criterium that I actually would expect the commission to abide by but I was curious to see if I could make it work given that it was mathematically possible (and I ended up being pretty happy with the final result).

Keeping Battle Creek and Kalamazoo together is definitely a positive, as is keeping Bay City, Saginaw, and Midland together.

[President Punxsutawney Phil]

Of course, I still like my map (https://districtr.org/plan/96349) more than any of these. That said, these final options are far superior to some of the absolutely hideous options that the commission was pondering earlier on. (It was also quite predictable that they would rather go for 2 majority-minority districts than take my approach, even though mine should still be totally VRA-compliant, so I'm not surprised about that.) I wouldn't be too upset with any of them, even though the Kalamazoo-GR district in the Apple plan really is quite dubious and Wahlberg's district reaching both Lake Michigan and Lake Erie is not really ideal as far as compactness goes.

https://davesredistricting.org/join/75dcd34d-39e2-4ff4-b049-bcc9d3284f42
thoughts on this map?
It's inspired by your map (which solves some problems I have had difficulty with), while also avoiding a whole-Detroit CD. It takes a less dogmatic approach to county splits than I usually do.
It also keeps Battle Creek and Kalamazoo together (not sure how much of a positive that is).

Looks really good! I had personally gone into the drawing process with the idea of keeping every single municipality, including the city of Detroit, undivided, which was not at all a criterium that I actually would expect the commission to abide by but I was curious to see if I could make it work given that it was mathematically possible (and I ended up being pretty happy with the final result).

Keeping Battle Creek and Kalamazoo together is definitely a positive, as is keeping Bay City, Saginaw, and Midland together.

It's really quite remarkable how Genesee, Bay, Saginaw, and Midland, combined, creates a district less than 10,000 away from quota.
Also, that 8th I drew is very much a district that PVI makes look vulnerable, but is solider than expected when you look at Biden numbers.

[Horus]

Hopefully we don't lose Andy Levin.

[GALeftist]

I'm torn; on the one hand, I think it makes the most sense for Grand Rapids to have a Biden district and Macomb to have a Trump district, but on the other hand I think pairing Grand Rapids with Muskegon and especially Kalamazoo is pretty dumb. Overall I would be happy with either Birch or Chestnut.