US House Redistricting: Ohio

[muon2]

If I assume GOP control, then this was my attempt to maximize their result. I kept counties as intact as possible and kept districts defensibly compact. The VRA district links Akron to Cleveland along the Cuyahoga Valley NP. Based on the nearly even presidential results of 2004 to judge the districts this would be 12-4 in favor of the GOP.

[muon2]

13 and 2 also could well go Democratic with such large portions of Cincinatti and Cuyahoga.

A Generic R could easily hold 2, given that Eastern Hamilton County is more Conservative than the rest.  The Current one is R + 13, and the one drawn here is not much different.

Actually both CD 1 and 2 as I drew them would have voted McCain in 2008. CD 1 is closest at 51.5-48.5, and CD 2 would have been 54-46. In an even statewide year like 2004 they both would be about 56 or 57% R.

[muon2]

13 and 2 also could well go Democratic with such large portions of Cincinatti and Cuyahoga.

A Generic R could easily hold 2, given that Eastern Hamilton County is more Conservative than the rest.  The Current one is R + 13, and the one drawn here is not much different.

Actually both CD 1 and 2 as I drew them would have voted McCain in 2008. CD 1 is closest at 51.5-48.5, and CD 2 would have been 54-46. In an even statewide year like 2004 they both would be about 56 or 57% R.

What's your approximation for CD-6 & CD-7.  They both seem to have the largest chunks of Franklin county.

They are fairly close and probably would have been for Obama by 51-49 or so, but would be R in most years. With precise precinct data I could adjust that by shifting Champaign (and maybe Logan) to 7 and bringing 4 deeper into Columbus. CD 8 could move into Springfield if needed and give up some R areas to CD 6.

[muon2]

13 and 2 also could well go Democratic with such large portions of Cincinatti and Cuyahoga.

A Generic R could easily hold 2, given that Eastern Hamilton County is more Conservative than the rest.  The Current one is R + 13, and the one drawn here is not much different.

Actually both CD 1 and 2 as I drew them would have voted McCain in 2008. CD 1 is closest at 51.5-48.5, and CD 2 would have been 54-46. In an even statewide year like 2004 they both would be about 56 or 57% R.

What's your approximation for CD-6 & CD-7.  They both seem to have the largest chunks of Franklin county.

They are fairly close and probably would have been for Obama by 51-49 or so, but would be R in most years. With precise precinct data I could adjust that by shifting Champaign (and maybe Logan) to 7 and bringing 4 deeper into Columbus. CD 8 could move into Springfield if needed and give up some R areas to CD 6.

Youve drawn a map where Democrats could win OH-01, OH-06, OH-07, OH-09, OH-10, OH-14, OH-13, OH-16, and probably OH-02 if Schmidt is still around in a good Dem year.  Republicans are going to have to concede a seat to Democrats in Columbus to keep the surrounding districts at least 54%-55% Republican.  

CD 9, 10 and 16 are strongly D and with CD 11 make up the 4 certain D seats in this map. As I noted, CD 1 and 2 would have voted for McCain, so it would take an even larger Dem wave than 2008 to topple the seats. I agree that as drawn CD 6 and 7 are vulnerable, but both are close with 2008 data. I'll post an update that can keep them R in that year.

[muon2]

Hey, Muon2, are you going to drink the kook-aid and now put all those Dems in Columbus in one of more GOP districts, rather than just give up, and give them a CD, or, alternatively do something creative, and combine them with some Dems in Cleveland or Akron, thereby creating some hideous looking gerrymander for the ages map, or what? 

What? You didn't appreciate my Sept. offering? Look how neat and compact most of the districts are as well. Of course, with the 5 district pick up this week, a map like this would still cost on GOP member in 2012.

If I assume GOP control, then this was my attempt to maximize their result. I kept counties as intact as possible and kept districts defensibly compact. The VRA district links Akron to Cleveland along the Cuyahoga Valley NP. Based on the nearly even presidential results of 2004 to judge the districts this would be 12-4 in favor of the GOP.

[muon2]

BTW I have a tough time believing that map would ever stand a VRA challenge since it is largely based on diluting the black population of Columbus as much as possible.

I don't recall now if it is legal to slice and dice the black vote, if there is not enough of it, which if put together, would make for a majority-minority "community of interest" CD.  I think there are something like 250,000 blacks in Franklin County. Right now, if I recall, there are mostly in Tiberi's CD.  That is why I wondered about the partisan lean of CD-6 on the map above, because Tiberi's district having checked it out, takes in the northeast corner of Franklin.

The largest Columbus-area district that is at least 50% black has only about 340 K people with 170 K black. That is only about half the population of a CD, and under Bartlett, the state has no obligation to maximize the black vote in that district. OH has no community of interest law that applies to CDs either. So, there should be no basis to challenge the split in that map.

I drew CD 6, 7, and 12 to all be about R+1 to R+3 using 2004 voting. Without a precinct map its hard to be more precise. With more precise data CD 4 can be used to improve the R performance of the other 3 Franklin districts. In a wave D year they all would be at risk, but the goal was to maximize GOP chances and hope that the incumbents hold it in a bad year. LaTourette is an example of that type of OH incumbent in 2008.

The incoming class is 13 R - 5 D, and OH will likely lose 2 seats. To even hold 12 R after redistricting in 2012 will require quite a few swing districts with only slight GOP lean.

[muon2]

The current map is already an obscene Republican gerrymander - if they tried to make it any more Republican, they would either run afoul of the VRA or set themselves up for a big reversal if the natinonal environment changed again - which was what happened in 2008.

The GOP challenge in OH is how to deal with the loss of two seats on the current map. Realistically they may have to reduce both one D and one R seat, but the remaining seats can be just as strongly drawn R as they were 10 years ago.

[muon2]

For a map to be real, one must know the Bush 2004 election returns, and do precinct by precinct analysis when one is divvying up counties, particularly the larger ones. Anything less than 54.5% Bush 2004 becomes vulnerable, and I try if possible to get to 55%, before I am satisfied. That would be a GOP PVI for the Bush 2004 numbers of +3% to +3.5%. That is what is needed for GOP incumbents to be reasonably safe if not flawed in a Dem year. And I won't draw a more heavily GOP district to save some weak incumbent who under performs, like say Bachmann, if it is going to make another GOP slated seat too vulnerable. Do you guys have different standards than that?

And this only obtains to areas north of the Mason Dixon line in general, that are not heavily Hispanic. When either of those factors obtain, it is a whole new ball game (for example, when might the Hispanics who are not voting now, start to vote?), and of course for Texas, the Bush numbers are essentially worthless, because they are inflated.

I have some of that voting data, and I am reworking my preelection map to see how well a 12-4 map could be constructed with new incumbents. SW OH is easy to get everyone up to 55% R without a lot of shifts. With a four-way split 53% R looks possible for the Columbus area, and I'm still looking to see if it can get higher.

NE OH does not look hopeful for such strong districts. I'm convinced that no splitting of Akron works for the GOP, and so it will have to be attached to one of the Cuyahoga districts to get it out of any intended R districts. Even so, LaTourette's district may not be better that 51% with '04 numbers. The new members in 6 (Johnson) and 18 (Gibbs) both live near the northern ends of their districts pressing more members into NE OH than before.

Johnson lives just south of Youngstown on the border of the district, and its hard to see how his home avoids being in a Youngstown-Warren district built for Ryan. That would fit with a natural inclination to link the heavily Dem areas in current CD 6 with Ryan's base. I also see that Renacci lives quite close to Sutton, which creates the possibility of putting them both in the same district with a 56% R vote in 2004. The GOP could then claim some fairness by eliminating one district from each party and setting up two general election matchups between incumbents, with Dems favored in one and the GOP favored in the other.

[muon2]

Ohio 12-4




I used the spreadsheet from the Ohio SoS site for county splitting.



Major changes:


CD-1 (Chabot) - I cut this from 36% black to 21% black, replacing them with Republicans from Claremont County

CD-2 (Schmidt) - Weakened a bit to shore up CD-1. Added Scioto and Lawrence Counties.

CD-3 (Turner)  - Adds a bit of heavy Republican territory to the east. I also used John Boehner to shore him up in Dayton a tad. If Mr. Speaker nixes that idea, well, Turner will have to take Trotwood back, but he's ok.

CD-4 (Jordan) - If he's not running for Senate, he comes back to a slightly weaker, but easily holdable district. The sections of Lorain county in his district actually voted ~54% McCain. The sections of Cuyahoga he has are ~50% McCain.

CD-5 (Latta) - There's a bit of an excess of Republicans here, but you can't do much with them.

CD-6 (Gibbs/Johnson) - This district is probably 50/50. There's really not much to do to make it safe; I think the GOP has to live with this one.

CD-7 (Austria) - About the same. Adds Democratic Athens County and some other Republican counties.

CD-8 (Boehner) - See above. Mostly unchanged.

CD-9 (Kaptur) - Adds heavily Democratic areas of Lorain County.

CD-10 (open) - Packed in Columbus seat. Safe Dem, no incumbent. I didn't really work on the specific borders that much.

CD-11 (Fudge) - Packed in Cleveland seat. Probably one of the most Democratic seats in the country outside of California. 50% black as I have it.

CD-12 (Tiberi) - He also staircases up to Stark County.

CD-13 (Ryan) - The old 17th, plus all of Akron and Cuyahoga Falls.

CD-14 (Latuorette) - Tricky. You can't take him west, so he goes South. The areas I added from Summit county (Tallmage, Richfield) are about 50% McCain. Boehner really needs to keep Latuorette in the House.

CD-15 (Stivers) - He doesn't live here, but the new CD-15 staircases up to Stark County.

CD-16 (Renacci) - Drops Stark County. Adds ~40% McCain areas of Cuyahoga County (Brooklyn, Parma). I did give him some Richland territory to compenate, but this district might have to swap areas with the 4th a bit.


If the GOP wanted to get really ugly, they could work CD-5 into Cuyahoga county by running a tendril through Lorain County.

Overall I can see 16 and 6 being vurnerable, but everything else looks fairly ironclad. Mean Jean might lose the 2nd, but a better Republican will probably win it back.

I activated the map. BTW Gibbs lives in Holmes county so that puts him in CD 15 in your map.

[muon2]

Thank you, muon2.

I did some rough math.

CD-16 is sitting at about 50/50. That's better than his current district but not really safe. Jim Jordan's CD-4 is about 56% McCain, I am not sure whether we can weaken that any more than I have. He's not exactly moderate.

CD-14 is also sitting at about 50/50, but that can't be helped.

CD-6's full counties are sitting about 53% McCain combined. The areas in Portage/Stark drag that down, though. To shore this district up, I think you can throw the town of Alliance into the 12th.

Every other district should be sitting about about 54% McCain or higher.

Don't overlook Sutton in your map. She lives right on the border of 14 and 16 as you've drawn it, and both districts are vulnerable to a Dem challenge. I think she would likely run in 16 since LaTourette is more entrenched in 14. In that case she would stand a good chance of winning in 2012, which I doubt would be the GOP's plan.

That's why I suggest making the district that includes Renacci much stronger than the other GOP districts in NE OH. You've already put Stivers or Gibbs out of a district so something has to be done to prevent the GOP conceding both seats in the reduction to 16 districts.
 

[muon2]

With the new ACS estimates, my OH-11 is under by 50,000 voters. I don't think there's going to be any way to keep Fudge's district majority-black.

If you connect CD 11 to Akron through Twinsburg a 54% black district is possible. CD 16 then connects Youngstown/Warren to Canton/Massillon and Kent. CD 14 remains pretty much as is.

[muon2]

It's interesting that you guys are able to get the Dems down to 3 districts in the NE. Imho that is even more of a reason to give the Dems a seat in Columbus and make sure you don't lose any central or western Ohio seats even in a bad year. It would really suck for the pubbies if a few districts flipped in the northeast as well as Columbus. And the trend is in the wrong direction in Columbus for the GOP.

Thus my rather extensive post about Columbus. It needs a lot of work to mitigate the damage. Otherwise the map is just an exercise in foolishness, in the longer term.

By the way, it is not just 3 Dem CD's in NE Ohio. It is 3 CD's in the entire state! We have the OH-09 snake with a double prong at the end, like that instrument you use to shove stuff around in your fireplace, and then the black - white liberal pack (OH-11), and the down and out quite white pack, but 20% black as it sucks up Dem precincts (OH-06).  The map is just a monster Gerry. It's just brutal. But we live in brutal partisan times - so the Pubbies might just do it.

I've had some intel that suggests that something like Torie's plan is exactly what the GOP is looking at. There must be a 50% black-VAP CD, and it will link Cleveland to Akron, since it can't be created in Cuyahoga alone. The map will also have a Toledo to Cleveland CD and one other Dem CD based in Youngstown. Columbus will be split into wedges.

The only major difference is that they probably won't wrap LaTourette's district. They assume that he can handle an R+0 district, and there is more concern to boost Renacci to hold off Sutton.

[muon2]

Hastert didn't dictate the map in IL, that was a prime example of how politicians sometimes put their interests ahead of their party's interests.  Chicago Democrats were fine with a Republican incumbent protection plan that took away one non-Chicago Democratic seat because things were less polarized which meant that they gained some influence from having Hastert as Speaker.  Also, I doubt Cantor dictated the Virginia map, as he wasn't a top member of the House Republican leadership yet.  Remember, in PA they didn't go with the map national Republicans originally wanted, that could certainly happen here.  

I also suspect Schmidt would've lost to Hackett in the district Torie's drawn for her.

I didn't mean the 2000 Virginia map.

Cantor, Goodlatte, and Connolly pretty much wrote the upcoming Virginia map, not the 2000 one. At least that's what reports are saying.

As far as 2000 Illinois is concerned, you're right about the reason Chicago Democrats did it, but as I understand it Hastert was heavily involved in the actual lines. The 8th district for instance was designed to protect Phil Crane although he lost it anyway, even while W was winning there easily.

Hastert's staff did draw the 2000 map for IL in consultation with Cong. Dan Lipinski. The Dems got the map they wanted in Cook and protected Cong. Evans in IL-17, but conceded the rest of the state. The legislature didn't necessarily like it, but the Speakership matters a lot. It will be interesting to see what role it plays in OH this year.

[muon2]

Jordon (OH-04) in my map just sucked up in a gratifyingly efficient manner (a nice little prong to the SE from his SE corner of his new CD), about 90,000 residents from Stivers' CD, few of which are in precincts that are  more than about 30% McCain, and most heavily black. We shall see if he can take the hit, but the balance of his CD basically hates Dems, so I suspect he can. It was an elongated sliver of course into Stivers' CD, leaving everything else in his old CD (what is left of it) in place. Now Stivers' CD will jut north like a knife to get the population he needs. Franklin has its precincts well organized in spreadsheet form, so I should be able to do all of this with some alacrity, without missing a precinct in the sense of putting it in the "wrong" CD, as I "see" what is "right" and "wrong."

Your OH-4 is a model for a mathematical redistricting paper I read recently. The statistical analysis concluded that it was best to pair extremes from both parties with a slight edge to the mapper's side to have the most effective gerrymander. Nominally that should be more successful than pairing and extreme R area with a moderate D area or vice versa.

[muon2]

Some mathematical analyses I've read suggest that it's best to put the really hard D areas in with the hardest R areas. Hard D areas will never cross over for a primary challenge, and as long as there are enough hard R areas, then the D can't win a general election either. If that analysis is correct, the maybe Athens should go to OH-2.

[muon2]

Some mathematical analyses I've read suggest that it's best to put the really hard D areas in with the hardest R areas. Hard D areas will never cross over for a primary challenge, and as long as there are enough hard R areas, then the D can't win a general election either. If that analysis is correct, the maybe Athens should go to OH-2.

You don't think a Bachmann/Schmidt would suffer disproportionate erosion in a university town, vis a vis a Portman, or Rogers, or Ryan, or even say my congresscritter, Campbell, at least to the extent you are not talking about a hard left University? Particularly in these times, when the status quo is really, really f'ing the young, and all we need is the right messenger to get the truth out? Folks from Athens won't be getting those prestigious federal jobs, the number of which has nearly doubled under Obama, and late term Bush. So they don't have that incentive. Most of them are destined to be middle to lower middle class "slugs," to put it brutally, but honestly. Does that make any sense to you Muon2?

The theory is that if there is a pool of hard D votes, there's nothing that can be done to win them in any year. Likewise a pool of hard R would never be lost, so the best district for a hard R candidate would have 51% hard R and 49% hard D with no independent or swing voters.

The swing voters should be reserved for districts that will attract candidates with a moderate appeal, and that would be where the hard core elements of the party can't win in the primary. Ideally it's a district with lots of swingy voters, but with a clear R PVI if the GOP is drawing the map.

This is just one theory, but it make some sense from a statistical perspective. It requires thinking about one's partisans as to the strength of their convictions. That's harder to discern from simple precinct results, but comparing strong elections for opposite parties (eg 2006 vs 2010) can provide a good sense of the amount of swing in a precinct.

[muon2]

Some mathematical analyses I've read suggest that it's best to put the really hard D areas in with the hardest R areas. Hard D areas will never cross over for a primary challenge, and as long as there are enough hard R areas, then the D can't win a general election either. If that analysis is correct, the maybe Athens should go to OH-2.

You don't think a Bachmann/Schmidt would suffer disproportionate erosion in a university town, vis a vis a Portman, or Rogers, or Ryan, or even say my congresscritter, Campbell, at least to the extent you are not talking about a hard left University? Particularly in these times, when the status quo is really, really f'ing the young, and all we need is the right messenger to get the truth out? Folks from Athens won't be getting those prestigious federal jobs, the number of which has nearly doubled under Obama, and late term Bush. So they don't have that incentive. Most of them are destined to be middle to lower middle class "slugs," to put it brutally, but honestly. Does that make any sense to you Muon2?

The theory is that if there is a pool of hard D votes, there's nothing that can be done to win them in any year. Likewise a pool of hard R would never be lost, so the best district for a hard R candidate would have 51% hard R and 49% hard D with no independent or swing voters.

The swing voters should be reserved for districts that will attract candidates with a moderate appeal, and that would be where the hard core elements of the party can't win in the primary. Ideally it's a district with lots of swingy voters, but with a clear R PVI if the GOP is drawing the map.

This is just one theory, but it make some sense from a statistical perspective. It requires thinking about one's partisans as to the strength of their convictions. That's harder to discern from simple precinct results, but comparing strong elections for opposite parties (eg 2006 vs 2010) can provide a good sense of the amount of swing in a precinct.

Doesn't this analysis kind of ignore turnout, though? For example, your "ideal" district of 51% hard R's vs. 49% hard D's would swing as soon as the Democrats had a turnout increase relative to the Republicans.

Of course there are many other factor that one needs to account for, including turnout. However, I thought the theory provided an interesting insight into districts that is not often covered. I would say that turnout is more likely to be a problem with swing voters. The hard D's and R's are usually the most likely to vote - that's part of how they get identified as such.

[muon2]

Exactly. Unless there is a state law provision against using water contiguity, you can use it (at least no court that I know of has said that you can't). People might complain about using water contiguity just for partisan gerrymandering, but it doesn't look any worse than the way you currently have the 16th District.

I suspect there may be case law on that matter. The GOP had control in 2001 and didn't hop across Erie except to islands or over bridges.

[muon2]

Jim Jordan's hijinks as head of the Republican Study Committee may lose him his seat.

Edit: I see this was already posted in the Ohio redistricting contest thread for some reason.

I've consolidated it all back here. That leaves the contest thread for the contest.

[muon2]

Right, I put him in Stivers' district (which was, in fact, my intention). Duh. Stupid rectangular counties.

Welcome to the Midwest! 

With homage to the Land Ordinance of 1785 and the Northwest Ordinance of 1787.

[muon2]

Here's my good government alternative. It's based on the Ohio Redistricting Competition and has exact equality at the block level. The minority district (11) is 48% black VAP, which has been agreed upon by the Ohio NAACP. The districts are highly compact with few splits.

The interesting feature is the competitive nature of the districts. Obama carried 11 of 16, but 13 of 16 went R in an average of 2010 races. One of the McCain districts (6) was only 49.7% R, so it could easily elect a D. Two of the 2010 D districts (13, 16) were carried by Obama only 4.6% and 2.7% ahead of his national average, so they could go R with the right candidate.

Effectively 11 of the 16 districts could flip on any given election with a maximum of 15 seats for the Rs and 12 for the Ds. More details can be found here.

[muon2]

Why is the 3rd drawn as it is?

To keep it from being too R, it needs to add Springfield and lose a heavy R suburb like Beavercreek. One could probably have gone up to Kettering on the Montgomery side as well.

[muon2]

Here's my good government alternative. It's based on the Ohio Redistricting Competition and has exact equality at the block level. The minority district (11) is 48% black VAP, which has been agreed upon by the Ohio NAACP. The districts are highly compact with few splits.

Is this confirmed?

I don't know if this will be the legal standard, but it is now the standard for the competition, which is intended to produce maps for consideration by the legislature. When it became clear that a 50% district was not possible without going into Akron, the competition revised the rules to allow 48% BVAP districts. The footnote indicates that this is considered to be sufficient to elect a black candidate of choice in Cuyahoga county at the federal level. I know that NAACP, who is a partner with the competition, was consulted and they agreed to the rule change.

[muon2]

Oh, it is a game. What was the NAACP signing off at 48% about? I saw an article about you Muon2 on this contest stuff.  Are you going to post it? You also bootstrapped off it I see. For some reason, you don't like the new Illinois lines much.

There were several articles about muon's win in the Columbus Dispatch.  I really liked his plans for the Ohio legislature. 

I'm not sure I like his congressional plan above as well though.  The competition software rates 4 of his districts with a GOP index of over 61% while there is only one district with a Democratic index that is over 54%.  That's great for competitiveness but I feel like the Democrats are sacrificing more for that cause than Republicans under that particular map.  I think you need to boost the Dem rating of at least one more seat to make it fair.

The competition website has been including links to articles about my plans. I hadn't duplicated those links here.

The problem with the congressional map from the Dems' perspective is that even a 48% BVAP district results in an 80%+ D district.  Since the state has a slight R lean to begin with, each additional hard D over 58% would knock out two or more lean D districts.

The map I posted has 7 lean D seats to create a 7R, 8D, 1E map adding leaners to the hard districts. A plan with two hard D districts, and no additional hard R districts would end up with something like a 10 R,  6D map or even an 11 R, 5 D split. That would be less fair compared to the statewide results, though it might look good to a Dem after the 2010 results alone.

[muon2]

In an interesting development, I was informed Fri that the OH House committee on State Government and Elections would hold a hearing on Sep 6, to receive testimony on congressional redistricting. The contest sponsored by the Ohio Campaign for Accountable Redistricting isn't over until 9/11, but they want to get something on record for this week's hearing. So, they will show the highest scoring map to date which is based on minimizing county splits and was linked here.

This "fair min splits" plan scores higher than my exact population plan, because it doesn't have to split counties to get exact equality. Like the exact plan it is based on a 4 SR / 3 LR / 1 TU / 7 LD / 1 SD mix, with 11 competitive districts out of 16. The only city split within a county is Columbus. Cleveland, Cincinnati, Akron, and Dayton are kept intact. Here's the rendering in DRA.