US House Redistricting: Michigan

[dpmapper]

Muon, making the graph of county splits a tree is exactly what I was going for with my metric for county splits.  I didn't state it in that language because I wasn't sure anyone would understand me.   

But that's about the only map I've seen that is a plausible gerrymander that nevertheless minimizes county splits both by my (sensible) standard and the 2001 standard simultaneously.  Nicely imagined. 

It seems that districts 12, 7, 8 might be a bit too marginal for the GOP's taste.  I think Rogers's district gaining not only Ypsilanti but also the Ann Arbor outskirts like Pittsfield and Scio Townships will outweigh losing East Lansing.  7 might be fine if not for its northern-most three cities.  And McCotter is kind of screwed. 

[dpmapper]

I've finished tabulating the numbers on the districts in my latest map.  

McCotter (light green) is still at 53.06% Bush, up slightly from 52.6% currently.  Not much can be done here, the way I have his district set up, but the good news is that voters here have probably been trending GOP the last 6 years.  

The two GOP districts that split Oakland are at a combined 55.5% Bush.  I didn't bother to calculate the intra-Oakland numbers but there must be a way to split it so that both stay at or above 55% Bush.  

Walberg's 7th (in grey) is at 54.67% Bush.  A bit lower than one would like it, and a bit lower than it is currently (54.97%).  Some of the areas outside Ann Arbor really hurt it.  With that area booming (by Michigan standards) this might be cutting it a little close.  Can't do much to help that without upsetting the entire apple cart.  

Upton's 6th (teal) is over 56% Bush.  Some almost-90% GOP townships in eastern Ottawa really work wonders.  

Camp's 4th (red) is at 54.90% Bush.  His part of Saginaw County actually helps matters, as it's at 56.5% Bush.  

Benishek's 1st (blue) is at 55.4% Bush.  

2nd (green) and 3rd (purple) will be quite safe, as even before adding in Kent County numbers, the 2nd is at 55.8% Bush and the 3rd is at 59.6%.  I won't bother with running the intra-Kent numbers; the mapmakers should feel free to split Kent in whatever way is most logical, rather than worrying about partisan numbers, as long as you put all of the minority parts of Grand Rapids into the 3rd (which would be the logical thing anyway).  

Haven't looked at Miller's numbers but she should be comfortable.  Her current district was over-Pubbied to get rid of Bonior.  

[dpmapper]

Is the goal here to hit the minimum number of county line breaks according to the current counting methods?  I see

2 in Wayne
1 in each of Oakland, Macomb, Kent, Ingham, Washtenaw, Saginaw, Genesee, St. Clair, Grand Traverse, and one TBA involving CD-06

for a total of 12, whereas the theoretical minimum is 9. 

[dpmapper]

Torie, you weren't paying attention to how Muon described the current (very dumb) counting procedure.  In the current map, since CD-09 is entirely contained in Oakland, it only counts as having 2 chops.  Wayne has only 1 chop currently.  The total is 11; efficiency 4.  In post #220 Muon describes how to get a chop count of 9 for 14 districts. 

I agree that the method for counting chops is stupid and ought to be changed irrespective of what gerrymanders it allows...

[dpmapper]

OK, you're down to 12 county splits by the current metrics.  That's the same as in my map in post 225, which has 4 Dem districts and every GOP district (save one) over 54.5% Bush.  

And sorry to continue to be a pain, but you have too many township splits in Wayne.  I see Sumpter Twp, Canton, Detroit, Taylor, and Dearborn Heights split, when you really should only need 3 splits between the 4 districts.  

Incidentally, you could improve things by exploiting the loophole that 1 whole district + 2 partials in a county is the same as 1 whole + 1 partial.  Give some parts of Oakland (Farmington, South Lyon) to CD-08 and let CD-11 take more of the Ann Arbor area. 

[dpmapper]

Muon laid out his interpretation of the rule in post 218:

You must be logged in to read this quote.

He then amended it very slightly in post #237 to deal with the case when the n pieces are not contiguous.  But that's it. 

I agree that if this is indeed the rule, it is stupid.  That is why I ignored the rule in my last map, and just went with the "normal" county fragment count.  But according to this method of counting, your map is at 12, which is the same as what I have. 

Regarding townships, I'd argue it's the same principle as counties.  In other words, splitting at most one between districts is a necessary condition towards satisfying the law, but not sufficient.  If there is a loop then there will be a way to lower the number of splits. 

For instance, there is a CD7-CD-14-CD8 circle of split townships.  This means that CD7 can take the rest of Dearborn Heights; CD 14 takes parts of Livonia rather than Canton (for contiguity purposes), and CD8 takes a little bit more of Sumpter.  That lowers the number of splits.  A similar loop exists between 13/14/7. 



More to the point, I don't see how your map improves on my map at all.  We both have the same number of county splits (by either Muon's (nonsensical) metric, or the "standard" metric) and my map's most marginal GOP district is quite a bit safer than yours. 

[dpmapper]

You MI-07 or MI-08 that pokes into Wayne is a Dem lean CD, maybe weak safe Dem CD. Your gray CD is at most but lean GOP. But you make both MI-12 and MI-10 considerably more GOP than my map in compensation, it looks like. Your map is a lot uglier than mine however, and that is not totally irrelevant. Cosmetics matter, and I pay considerable attention to that. You get ugly only when you really have to. Both our maps are a mess in the Detroit metro area of course.

The part-Wayne CD is at 53.06% Bush '04, slightly better than McCotter's current district.  As I mentioned earlier, I suspect that the southern tier of Wayne suburbs and Monroe County are trending GOP.  Monroe's Obama percentage was -1.6 from where he was nationally, whereas for Kerry it was +.4.  Nor did Dingell do well in these places last year.  The grey CD that takes part of Washtenaw is at 54.67% Bush, though it might be trending the other way.  See post #236 for more numbers. 

 I noticed your recent tallying has been using '08 numbers, whereas way back you were working with '04 numbers - why'd you switch?  

[dpmapper]

dpmapper, I am still playing with your map (subject to a delay, as I enrich turbotax with a multiplicity of tax returns that I am saddled with generating), but it appears more and more likely that you have drawn the best map that can possibly be drawn in its overall design, if not necessarily the assigning of every precinct, although in general you are spot on there as well. Your clearly know what you are doing, and are one very smart puppy!  The geography between Grand Rapids and Ann Arbor is such, along with county populations, that the options are very limited.

Awww, that's sweet of you.   

Michigan is a nice challenge - the statute makes it almost a topological exercise.  Other states are pretty simple in comparison - just find the bluest precincts and smash them together, more or less.   That's why I found it more interesting in PA to make extra conditions like "Bucks and Lancaster stay in one piece" - it puts concrete limits on the amount of ugliness tolerable. 

[dpmapper]

Huh, they're going for two Oakland/Wayne districts.  I guess they'll argue that they're forced to do so by VRA considerations. 

But if you can do two such districts, why not go for 9-4-1, rather than 9-5?  Pretty conservative. 

The interesting thing about this map is that they take Battle Creek and Mark Schauer out of Walberg's district.  Smart. 

[dpmapper]

Pretty sure the Sander Levin district drops to about 52-53% Kerry or so, and 59% Obama. It dropped its best portions.

It lost Southfield, but picked up Pontiac, and some marginally Dem areas in Macomb. It's Dem PVI is not going to drop that much. 

Look again, Torie - Pontiac is in the 14th.  The parts of Macomb it picked up are in Sterling Heights, so roughly even.  If the goal is to make 9 swingier, I'd tweak the border with MI-11 a bit.  MI-11 was already quite safe and they've made it safer by adding the rest of the thumb, so you could put Mt. Clemens and other parts of SE Macomb into MI-11 quite easily.  

[dpmapper]

And a zoomed in view of the Metro-Detroit area:

What on earth is going on with the 13th?  From Southfield to west Detroit and Inkster, but then bypassing Farmington Hills, Romulus and Westland to drop all the way down to Grosse Ile and northeast Monroe County?!

[dpmapper]

All this Lansing talk gave me an idea.
 



If the GOP gets super ambitious, and still feels they can get away with one extra county split with a black district going from Wayne to the north, and McCotter is retiring, then here's a 10-4, with all 10 GOP seats between 48.5 and 51.5% Obama.  

Blue Northern Michigan: 49.9-48.3 O-M (Benishek)
Lavender Midland/Bay City: 50.3-47.9 O-M (Camp)
Red Holland/Muskegon: 48.5-49.9 O-M (Huizenga)
Purple Grand Rapids: 51.3-47.0 O-M (Amash)
Teal Kalamazoo: 51.5-46.9 O-M (Upton)
Lime Green southern seat: 49.4-48.8 O-M (Walberg)
Green thumb: 51.2-46.9 O-M (open)
Bluish gray St Clair/Macomb: 51.4-46.8 O-M (Miller)
Pink Oakland County: 50.8-47.6 O-M (Knollenberg to challenge Peters)
Grey Livingston County/Livonia/southern Wayne: 51.3-47.0 O-M (Rogers)

Levin and Dingell are both drawn into the tan district which is 54% black; the brown district is 53.5% black.  The yellow Lansing/Ann Arbor seat is wide open. 

[dpmapper]

All this Lansing talk gave me an idea.
 



If the GOP gets super ambitious, and still feels they can get away with one extra county split with a black district going from Wayne to the north, and McCotter is retiring, then here's a 10-4, with all 10 GOP seats between 48.5 and 51.5% Obama.  

Blue Northern Michigan: 49.9-48.3 O-M (Benishek)
Lavender Midland/Bay City: 50.3-47.9 O-M (Camp)
Red Holland/Muskegon: 48.5-49.9 O-M (Huizenga)
Purple Grand Rapids: 51.3-47.0 O-M (Amash)
Teal Kalamazoo: 51.5-46.9 O-M (Upton)
Lime Green southern seat: 49.4-48.8 O-M (Walberg)
Green thumb: 51.2-46.9 O-M (open)
Bluish gray St Clair/Macomb: 51.4-46.8 O-M (Miller)
Pink Oakland County: 50.8-47.6 O-M (Knollenberg to challenge Peters)
Grey Livingston County/Livonia/southern Wayne: 51.3-47.0 O-M (Rogers)

Levin and Dingell are both drawn into the tan district which is 54% black; the brown district is 53.5% black.  The yellow Lansing/Ann Arbor seat is wide open.  

Now that McCotter is definitely going for Pres I revisited this and did a bit of rejiggering (such as finding a way to give the Flint district less of upper Oakland and more of Farmington Hills, and putting the GOP parts of Dearborn into the grey district).  The basic map is still there but the new percentages are

Blue Northern Michigan: 50.1 Obama (Benishek)
Lavender Midland/Bay City: 50.1 Obama (Camp)
Red Holland/Muskegon: 50.2 Obama (Huizenga)
Purple Grand Rapids: 50.6 Obama (Amash)
Teal Kalamazoo: 50.8 Obama (Upton)
Lime Green southern seat: 49.4 Obama (Walberg)
Green thumb: 51.2 Obama (open)
Bluish gray St Clair/Macomb: 51.0 Obama (Miller)
Pink Oakland County: 50.7 Obama (Knollenberg to challenge Peters/Levin)
Grey Livingston County/Livonia/southern Wayne: 50.8 Obama (Rogers, possibly Dingell?)

So they're all +2 or +3 R in PVI, not that ugly, almost-minimal county splits, and, apart from the red district, the lime green district, and (especially) the thumb district, pretty good as far as COIs go.  

[dpmapper]

I don't know... I think I'd rather have Conyers than Fieger.  *shudder*

[dpmapper]

That's because I didn't post the updated map.  Here's a pic of it:



No touch-point contiguity is needed for the Flint-Pontiac connection, even though it kind of looks that way.  There is one more county split between Wayne and Monroe/Oakland than is necessary, as in the real map. 

[dpmapper]

C'mon, Torie, where's the killer instinct?  Use water contiguity to take in Grosse Ile rather than Harper Woods! 

(Nice map.  I guess Mike Rogers must be really attached to Lansing, huh?) 

[dpmapper]

One could also argue that Livingston and northern Oakland are likely to grow with exurban population as well, I suppose.

You seem to have one too many city/township splits in Oakland Co, though - three split between three districts.  I think you need to only use two.