US House Redistricting: Ohio

[muon2]

45% invalid? Something stinks here. Possibly both ends.

Probably not. In large, statewide signature-gathering efforts, with untrained gatherers, there is often a very large number of invalid signatures. Duplicate signatures are common, since many people won't recall if they signed a particular petition, so they will sign again. Add in unregistered residents and non-residents, and conventional wisdom is that at least double the minimum should be turned in. Many recommend triple the minimum.

[greenforest32]

The additional batch of signatures has been submitted: http://www.therepublic.com/view/story/f10b80151194480fbc34d52084c99e07/OH--Ohio-Redistricting-Amendment

July 30, 2012
COLUMBUS, Ohio — A coalition pushing changes to the way Ohio draws legislative and congressional districts says it now has enough signatures to qualify its proposed constitutional amendment for the fall ballot.

The Voters First coalition fell more than 130,000 shy of the roughly 385,000 valid signatures needed to appear on the ballot. It also failed to meet certain county requirements.

The group's leaders on Monday said they filed almost 301,000 additional signatures with state officials this weekend, and they are confident they now meet the requirements.

[Senator North Carolina Yankee]

When will we find out if they have actually got it done?

[greenforest32]

Probably in 10 days or so. If it made the ballot, I don't know if it would pass. Redistricting-related initiatives have failed before: http://ballotpedia.org/wiki/index.php/Redistricting_measures_on_the_ballot#tab=By_year

[Senator North Carolina Yankee]

If it fails, they should try again with one that takes effect in 2020. It is hard to get these passed when the dominant party is 100% hostile, especially if the minority party is lame in its support.

How much effort are the OH Dems putting into this?

[Gravis Marketing]

If it fails, they should try again with one that takes effect in 2020. It is hard to get these passed when the dominant party is 100% hostile, especially if the minority party is lame in its support.

How much effort are the OH Dems putting into this?

I don't know, but they were AWOL on the last effort...

[Senator North Carolina Yankee]

If it fails, they should try again with one that takes effect in 2020. It is hard to get these passed when the dominant party is 100% hostile, especially if the minority party is lame in its support.

How much effort are the OH Dems putting into this?

I don't know, but they were AWOL on the last effort...

Which makes no sense unless they desire to have their crack at it for revenge purposes next time and are willing to sacrifice this decade.

[Gravis Marketing]

If it fails, they should try again with one that takes effect in 2020. It is hard to get these passed when the dominant party is 100% hostile, especially if the minority party is lame in its support.

How much effort are the OH Dems putting into this?

I don't know, but they were AWOL on the last effort...

Which makes no sense unless they desire to have their crack at it for revenge purposes next time and are willing to sacrifice this decade.

I think it's just incompetence and lack of common focus in the party. The Ohio Dems are a sad lot.

[Mr.Phips]

Democrats really need to sweep the statewide offices in 2018, which would lead to favorable legislative legislative district lines through the reapportionment commission, giving them likely control of the state House and maybe even the state Senate after 2022. 

[greenforest32]

The measure has qualified for the November ballot: www.vindy.com/news/2012/aug/06/ohio-redistricting-issue-headed-november-ballot/

COLUMBUS — A group hoping to remove politics from the way Ohio draws its congressional and legislative district lines has collected more than enough valid signatures to place the issue on the November ballot.

Late today, Ohio Secretary of State Jon Husted certified 406,514 registered voters names submitted by Voters First, above the 385,000-plus required to qualify for the general election. The group also met required threshold levels in 60 counties.

It would redraw the existing maps if it passed: http://www.cleveland.com/open/index.ssf/2012/08/ohio_elections_chief_validates.html

If approved by voters, the plan would go into effect immediately so that the commission would draw a new map in 2013 in time for the 2014 election.

[RBH]

Map in case of the amendment passing this fall and new lines coming up for 2014

First 8 districts numbered by how far south they are, Last 8 numbered by how far north they are (with the last 5 numbered from West-East on Lake Erie)

Statewide:


District 1 (open, Wenstrup?):

59/39 McCain, 52/48 R, 95W/2B

District 2 (Chabot and Wenstrup):

55/44 Obama, 52/48 D, 65W/28B

District 3 (Boehner):

64/35 McCain, 63/37 R, 87W/5B

District 4 (Bill Johnson or Charlie Wilson):

49/48 Obama, 60/40 D, 94W/3B

District 5 (Jordan):

59/39 McCain, 58/42 R, 89W/6B

District 6 (Turner):

51/47 McCain, 53/47 R, 78W/16B

District 7 (Tiberi and Stivers):

52/47 McCain, 54.5/45.5 R, 81W/8B

District 8 (Joyce Beatty):

65/33 Obama, 64/36 D, 62.5W/27B/5H

District 9 (Gibbs):

58/40 McCain, 55.5/44.5 R, 92W/4B

District 10 (open, Tim Ryan? Joyce Healy-Abrams?):

55/43 Obama, 63/37 D, 85W/9B

District 11 (Sutton or Renacci):

55/44 Obama, 60/40 D, 83.5W/11B

District 12 (Kaptur):

56/42 Obama, 58/42 D, 79W/12B

District 13 (Latta):

51/47 Obama, 54/46 D, 87W/4B/6H

District 14 (open):

55/44 Obama, 60/40 D, 87W/5B

District 15 (Fudge):

80/20 Obama, 80/20 D, 42W/49B (45.5 VAWhite-46.2 VABlack)

District 16 (David Joyce and Tim Ryan):

52.5/46 Obama, 58/42 D, 91W/4.5B

A look at Cuyahoga County:


And my first draft (where I decided to switch things up for 5, 7, 8, 9 and 13):

Figured 8 and 13 were not particularly concise, and did some redrawing over those 5)

Overall
x Safe R (1, 3, 5, 9)
x Lean R (6, 7)
x Tossup (4)
x Lean D (2, 13, 16)
x Safe D (8, 10, 11, 12, 14, 15)

Thoughts?

[muon2]

Of course if the referendum passes and the weighting factors match the 2011 competition I would be partial to the map on the right. It is the one filed by the Dems last fall and has 4 strong R, 3 lean R, 1 even, 7 lean D, and 1 strong R. The leans are all within 10% differential between the parties and most are within 5% (52.5% to 47.5%)

[Vasall des Midas]

...and of course, it was drawn by a politician.

[muon2]

...and of course, it was drawn by a politician.

Of course.

[TJ in Wisco]

I like the competition proposal map for the most part with a couple caveats:

1. The Lake/Geauga to Trumbull district looks nice on paper but is really ugly in reality because most of the people are concentrated at the ends of it in two different metro areas. Unfortunately this is a difficult one to draw no matter what way you go, but this way you end up splitting the Mahoning Valley metro area pretty badly. It looks clean on a map and clean by county lines but it isn't actually clean.

2. Columbus is a blatant competitiveness gerrymander. The Columbus CD leaves out areas close to the city center in the City of Columbus to instead cover outlying suburbs. 7 and 13 need to loose the arms into each other. 13 should start in Downtown Columbus and first cover neighboring areas within the city before moving on to outer suburbs.

3. The Dayton area is a Democratic gerrymander. If CD 8 needs to take an arm into part of the Dayton metro, the drawers should at least pretend to make it shaped reasonably instead of an arm right through Beavercreek. CD 3 needs to loose the extraneous arm into Springfield. Springfield is not part of the Dayton metro area, which doesn't mean it can't be in the Dayton CD, but the entire CD shouldn't be configured just to find a way to put Springfield into it.

[Torie]

Yes respecting metro areas trumps county lines any day of the week. I agree with TJ on that.

[muon2]

Metro areas are not part of the OH referendum. IMO they are only slightly removed from the squishy subject of communities of interest. The referendum amendment would consider county and municipal lines. It also will consider competitiveness and representational fairness and how that is weighted can drive the choices in a map. They were given equal weight with geography in the competition and that resulted in all the specific choices that TJ notes.

[TJ in Wisco]

Metro areas are not part of the OH referendum. IMO they are only slightly removed from the squishy subject of communities of interest. The referendum amendment would consider county and municipal lines. It also will consider competitiveness and representational fairness and how that is weighted can drive the choices in a map. They were given equal weight with geography in the competition and that resulted in all the specific choices that TJ notes.

I think you are showing just how a competitiveness criterion acts as a defacto Democratic gerrymander here. In order to meet it, the map drawers will gerrymander certain areas, like the places I noted, by choosing adjacent areas with much less geographic connection in order to turn what would otherwise be Republican-favoring seats into competitive ones. This is done in an obvious way in CDs 3,7, and 14 and to a lesser extent with CD 9 (though Lorain County doesn't really fit well anywhere).

Oh, and one other thing, the map drawers really need to pick which corner of Hamilton County to chop off and stick with it. The tendril going down the eastern side is a bit ridiculous.

[RBH]

when it comes to the map I put up... if I just cut Franklin in two between 7 and 8, i'd probably create two attractive reliably D districts.

But the competition map splitting Franklin into 4 districts is a bit much.

There's really no way of creating an Ashtabula district without having it involve heavy populations on both sides... short of having the district go from East Cleveland into the PA state line.

Looking at the Ohio TV market map: http://dishuser.org/TVMarkets/Maps/ohio.gif

pretty much all maps openly mock TV market lines

[traininthedistance]

So, I didn't enter the contest from long ago, but had I done so, I would've gone with the following map.



The first thing this map does is identify a number of whole-county groupings, each centered around a metro area and nearby rural counties, which are close enough to a multiple of the idea CD population.  I have five such regions here:  Cincinnati/Dayton (1, 2, 3, ; Toledo (4, 9); Columbus (7, 12, 15); NE Ohio (5, 10, 11, 13, 14, 16); and Zanesville/Ohio River (6).  District 6 has the largest deviation, at 1,198, but it is also entirely whole counties; every other district is within 850 of the ideal, and could be brought to exact equality with microchops.

The second thing this map does is rigorously adhere to municipal boundaries.  Only three cities are split: Cleveland and Akron are split to allow a 50% BVAP VRA district; and Columbus is split because municipal boundaries in Franklin County are beyond insane and I give up.  Except for Hamilton and Cuyahoga, no county has more than 2 CDs; those two have three.

Obviously, with the high-level groupings as well as the within-group splits, I've attempted to keep metro areas as close together as possible.  There are a couple portions which are not entirely satisfactory: putting Madison in with the southeast group rather than Columbus is the worst offender, but as far as I can tell the 4-CD math made it hard to avoid.

Onto the districts!



1: Entirely Hamilton; Cincinnati and close-in suburbs.  55.3% Obama, 52.2% Dem.  Lean D.

2: Splits Hamilton with 1 and 8, and Greene with 3.  Western suburbs of Cincy/Dayton and rural southern Ohio.  37.3% Obama, 41.4% Dem.  Safe R.

3: Dayton/Springfield; takes Fairborn and Bellbrook from 2 to equalize population.  The decision to pair Dayton with Springfield rather than Beaverbrook/Xenia is pretty much the only thing Repubicans have any reasonable cause to complain about with this map IMO; but having 2 hook over to take Springfield would be uglier, and Dayton/Springfield is reasonable.  51.0% Obama, 50.5% Dem.  Tossup.

4: Lima, Findlay, and lots of farms in the northwest; splits Seneca with 9.  36.9% Obama, 39.8% Dem.  Safe R.

5: Centered on Elyria/Lorain, this entirely new compact collection of exurbs and small cities in northern Ohio is mostly within Cincy's orbit, but on its outer reaches.  Splits Lorain with 10.  50.6% Obama, 55.6% Dem.  Tossup.

6: Zanesville and Portsmouth are the largest cities here, I guess, but really it's a rural and Appalachian-flavored Ohio River valley district.  No splits whatsoever.  Ancestrally Dem, but Bill Johnson would probably be pretty happy with it, since I pushed it further south and away from Youngstown.  46.2% Obama, 58.1% Dem.  Lean R.



7: The southern Columbus-area district. Splits Licking with 12 and Franklin with 15; the city of Columbus itself is split, too, because it's impossible to get any sort of reasonable lines otherwise.  Also the deviation is down to -182; splitting Columbus helps get the numbers quite low here. Most of the more urban and liberal areas are in 15, but a few bleed into this district, which should still be R for now but is likely to trend more competitive.  47.8% Obama, 49.4% Dem.  Lean R.

8: Boehner's district is still mostly the northern suburbs of Cincy with some rural areas to fill out population; it sensibly retreats from Dayton and the only split is in Hamilton with 1 and 2.  34.6% Obama, 36.7% Dem.  Safe R.

9: Toledo and environs; splits Seneca with 4.  59.2% Obama, 61.2% Dem.  Safe D.



10: I would have loved to keep this district entirely within Cuyahoga, but that wasn't compatible with a 50% BVAP district on Cleveland's East Side.  Pushing one township deep into Lorain is the next best thing, though.  Splits Lorain with 5 and the city of Cleveland with 11.  56.3% Obama, 61.6% Dem.  Lean D.

11: The VRA district does not quite maximize the black percentage, but it gets it over 50 with a minimum of ugliness and a maximum of useful ripple effects to neighboring districts (esp. 13).  53.3% black, 50.6% BVAP, splits Cuyahoga with 10 and 14 and Akron (both city and county) with 13.  80.3% Obama, 80.0% Dem.  Safe D.

12: North of Columbus, exurbs and rural areas.  Splits Licking with 7, deviation 189.  40.0% Obama, 43.0% Dem.  Safe R.

13: Akron (minus the parts taken for Fudge's district) and Canton!  These two counties seem to be a natural fit in my eyes, though maybe that's because they're linked in my mind due to being the birthplace of the National Football League, with its first champion (Akron Pros) and first dynasty (Canton Bulldogs).  52.4% Obama, 56.9% Dem.  Lean D.

14: LaTourette's district has no choice but to expand out to the borders of Cincy and Youngstown; it now splits Cuyahoga with 10/11 and Trumbull with 16 (though it takes nearly all of Trumbull).  53.5% Obama, 58.5% Dem.  I'll call this one Lean R, despite having better D numbers than 13, due to having such a popular incumbent.

15: The Columbus district.  Entirely within Franklin, takes in a couple northern suburbs and chops off the southern edge of the city to 7 in the name of sane lines and a -92 deviation.  63.2% Obama, 60.6% Dem.  Safe D.

16: Youngstown and similar areas to the south, such as New Philadephia and Steubenville.  Would've been nice to connect Warren, too, but 14 is already painted into a corner, so it eats into the old 6 instead.  Splits Trumbull with 14.  54.2% Obama, 66.0% Dem.  Safe D.

Final tally:
4 Safe R (2, 4, 8, 12)
3 Lean R (6, 7, 14)
2 Tossup (3, 5)
3 Lean D (1, 10, 13)
4 Safe D (9, 11, 15, 16)

Seems perfectly balanced to me.

[muon2]

So, I didn't enter the contest from long ago, but had I done so, I would've gone with the following map.



The first thing this map does is identify a number of whole-county groupings, each centered around a metro area and nearby rural counties, which are close enough to a multiple of the idea CD population.  I have five such regions here:  Cincinnati/Dayton (1, 2, 3, ; Toledo (4, 9); Columbus (7, 12, 15); NE Ohio (5, 10, 11, 13, 14, 16); and Zanesville/Ohio River (6).  District 6 has the largest deviation, at 1,198, but it is also entirely whole counties; every other district is within 850 of the ideal, and could be brought to exact equality with microchops.

The second thing this map does is rigorously adhere to municipal boundaries.  Only three cities are split: Cleveland and Akron are split to allow a 50% BVAP VRA district; and Columbus is split because municipal boundaries in Franklin County are beyond insane and I give up.  Except for Hamilton and Cuyahoga, no county has more than 2 CDs; those two have three.

Obviously, with the high-level groupings as well as the within-group splits, I've attempted to keep metro areas as close together as possible.  There are a couple portions which are not entirely satisfactory: putting Madison in with the southeast group rather than Columbus is the worst offender, but as far as I can tell the 4-CD math made it hard to avoid.


Final tally:
4 Safe R (2, 4, 8, 12)
3 Lean R (6, 7, 14)
2 Tossup (3, 5)
3 Lean D (1, 10, 13)
4 Safe D (9, 11, 15, 16)

Seems perfectly balanced to me.

Nice work. The regional approach is along the lines of what I worked on in CA, and what led me to the data set collected in my IA-style studies. It's probably too GOP as measured by the contest, and lacks the competitiveness to score lots in that category, but it would quite possibly have been in the top 10. If I can figure out the conversion from DRA to competition partisanship I'll give you a better assessment.

[muon2]

Here's some equivalent conversions for train's map from the competition plan I linked. The competition used two party votes from the 2010 statewide races that affected the composition of the apportionment board. I quote the GOP fraction, then the McCain part of the two-party vote and GOP part from DRA.

CD 1: Comp 47.52% McCain 44.4% DRAR 47.9%
CD 2: Comp 61.88% McCain 61.3% DRAR 54.2%
CD 3: Comp 51.83% McCain 47.9% DRAR 49.4%
CD 4: Comp 67.59% McCain 64.3% DRAR 61.1%
CD 5: Comp 48.49% McCain 44.0% DRAR 42.7%
CD 6: Comp 52.35% McCain 50.9% DRAR 41.1%
CD 7: Comp 48.67% McCain 44.6% DRAR 44.9%
CD 8: Comp 67.26% McCain 64.0% DRAR 63.0%
CD 9: Comp 49.35% McCain 44.7% DRAR 42.3%
CD 10: Comp 53.56% McCain 49.3% DRAR 44.8%
CD 11: Comp 20.22% McCain 16.1% DRAR 16.3%
CD 12: Comp 61.57% McCain 57.1% DRAR 55.7%
CD 13: Comp 46.80% McCain 42.7% DRAR 37.8%
CD 14: Comp 48.82% McCain 46.2% DRAR 42.3%
CD 15: Comp 47.72% McCain 42.9% DRAR 44.4%
CD 16: Comp 46.52% McCain 44.3% DRAR 38.5%

So the competition is about 2-4% more R than the McCain fraction which is slightly more R than the McCain to PVI shift of about 2.4%. The shift is quite dramatic compared to the DRA avg which is based on the 2006 Govs race.

[muon2]

So here's my analysis of the competition quality of train's map. I've used his Obama number with the likely 3rd party vote to get an R fraction then adjusted it using the table above. Note the competition is about 1.5% more R on average than an equivalent PVI.

CD 1: 47.2% Likely D
CD 2: 62.6% Safe R
CD 3: 51.4% Lean R
CD 4: 65.7% Safe R
CD 5: 53.0% Likely R
CD 6: 54.1% Likely R
CD 7: 55.5% Strong R
CD 8: 68.2% Safe R
CD 9: 44.7% Strong D
CD 10: 47.2% Likely D
CD 11: 23.1% Safe D
CD 12: 63.7% Safe R
CD 13: 50.8% Tossup
CD 14: 48.1% Lean D
CD 15: 40.5% Safe D
CD 16: 47.0% Likely D

Competition scores (with high score for each category)
Fairness leans 3.4% more R than the state as a whole: 86.4 points (top score 99.6)
Competitiveness 3 highly competitive, 5 competitive, 2 somewhat competitive: 21 points (top score 33)
County splits 18: 32 points (top score 43)
Compactness: not scored

If I shifted them by 1.5% to get a PVI, then the fairness zooms to 99.2 points and the competitiveness rises to 22 points. The choice of election data really matters in assessing fairness (cf AZ).

[muon2]

Here's the competition analysis for rbh's map:

CD 1: 60.6% Safe R
CD 2: 47.6% Lean D
CD 3: 68.8% Safe R
CD 4: 49.9% Tossup
CD 5: 63.3% Safe R
CD 6: 55.9% Strong R
CD 7: 57.1% Strong R
CD 8: 68.2% Safe R
CD 9: 63.1% Safe R
CD 10: 46.2% Likely D
CD 11: 49.6% Tossup
CD 12: 47.5% Lean D
CD 13: 52.7% Likely R
CD 14: 48.7% Lean D
CD 15: 24.2% Safe D
CD 16: 48.3% Lean D

Competition scores (with high score for each category)
Fairness leans 10% more R than the state as a whole: 60 points (top score 99.6)
Competitiveness 6 highly competitive, 2 competitive, 2 somewhat competitive: 24 points (top score 33)
County splits 26: 24 points (top score 43)
Compactness: not scored

With the 1.5% shift to PVI this goes to 85.6 points for fairness and 21 points for competitiveness.

[jimrtex]

This would have been my map based on no county splits outside the 3 largest counties, which must be split, then refining by shifting townships and precincts, with the intent of minimizing what might be described as displaced population, areas that are in districts other than most of the county population.

Some attention was paid to metropolitan areas and compactness.  No attention was paid to subjective political criteria.



NE Ohio


SW Ohio


Cuyahoga County


Franklin County


Hamilton County