Hypothetical NC Senate Map

North Carolina needs to draw new legislative maps. I expect legislators to draw something like this for the NC Senate. Currently, the GOP controls the upper chamber by a 35-15 margin.

Descriptions of the proposed districts, along with 2016 election data, are below. Feedback would definitely be appreciated! Please let me know if there’s anything that could enhance this map. Thanks!

1 – Camden, Chowan, Currituck, Dare, Gates, Hertford, Hyde, Pasquotank, Perquimans, Tyrrell, Washington
OPEN SEAT
+13% Trump
+12% Burr
+11% McCrory

2 – Carteret, Craven, Pamlico
Norm Sanderson (R)
+32% Trump
+33% Burr
+30% McCrory

3 – Beaufort, Bertie, Martin, Northampton Vance, Warren
Bill Cook (R) AND Erica Smith-Ingram (D)
+9% Clinton
+8% Ross
+8% Cooper

4 – Edgecombe, Halifax, Wilson
OPEN SEAT
+19% Clinton
+19% Ross
+22% Cooper

5 – Greene, Pitt
Don Davis (D)
+6% Clinton
+2% Ross
+6% Cooper

6 – Jones, Onslow
Harry Brown (R)
+34% Trump
+35% Burr
+33% McCrory

7 – Lenoir, Wayne
Louis Pate (R)
+9% Trump
+10% Burr
+10% McCrory

8 – Bladen, Brunswick, New Hanover, Pender
Bill Rabon (R)
+24% Trump
+25% Burr
+18% McCrory

9 – New Hanover
Michael Lee (R)
+6% Trump
+9% Burr
+3% Cooper

10 – Duplin, Johnston, Sampson
Brent Jackson (R)
+23% Trump
+23% Burr
+20% McCrory

11 – Johnston, Nash
Angela Bryant (D)
+15% Trump
+15% Burr
+10% McCrory

12 – Harnett, Johnston, Lee
Ronald Rabin (R)
+23% Trump
+23% Burr
+18% McCrory

13 – Columbus, Robeson
Danny Britt (R)
+11% Trump
+9% Burr
+11% McCrory

14 – Wake
Dan Blue (D)
+43% Clinton
+40% Ross
+43% Cooper

15 – Wake
John Alexander (R)
+40% Clinton
+31% Ross
+41% Cooper

16 – Wake
Jay Chaudhuri (D)
+28% Clinton
+19% Ross
+29% Cooper

17 – Wake
Tamara Barringer (R)
+3% Trump
+10% Burr
+0.2% McCrory

18 – Franklin, Wake
Chad Barefoot (R)
+8% Trump
+13% Burr
+5% McCrory

19 – Cumberland, Hoke
Ben Clark (D) AND Wesley Meredith (R)
+7% Trump
+10% Burr
+8% McCrory

20 – Durham
Floyd McKissick (D)
+77% Clinton
+71% Ross
+74% Cooper

21 – Cumberland
OPEN
+44% Clinton
+41% Ross
+40% Cooper

22 – Durham, Granville, Person
Mike Woodard (D)
+15% Clinton
+11% Ross
+16% Cooper

23 – Orange
Valerie Foushee (D)
+39% Clinton
+33% Ross
+39% Cooper

24 – Alamance, Guilford
Rick Gunn (R)
+16% Trump
+18% Burr
+10% McCrory

25 – Anson, Moore, Richmond, Scotland
Scott McInnis (R)
+15% Trump
+15% Burr
+11% McCrory

26 – Caswell, Rockingham, Stokes, Surry
Phil Berger (R)
+37% Trump
+37% Burr
+27% Cooper

27 – Guilford
Trudy Wade (R)
+9% Trump
+15% Burr
+2% McCrory

28 – Guilford
Gladys Robinson (D)
+61% Clinton
+59% Ross
+62% Cooper

29 – Guilford, Randolph
Jerry Tillman (R)
+33% Trump
+32% Burr
+25% McCrory

30 – Alleghany, Ashe, Surry, Watauga, Wilkes
Deanna Ballard (R) AND Shirley Randleman (R)
+37% Trump
+36% Burr
+26% McCrory

31 – Forsyth, Yadkin
Joyce Krawiec (R)
+28% Trump
+33% Burr
+20% McCrory

32 – Forsyth
Paul Lowe (D)
+44% Clinton
+39% Ross
+46% Cooper

33 – Davidson, Montgomery
Cathy Dunn (R)
+47% Trump
+47% Burr
+37% McCrory

34 – Rowan, Stanly
OPEN SEAT
+42% Trump
+40% Burr
+35% McCrory

35 – Union
Tommy Tucker (R)
+31% Trump
+35% Burr
+29% McCrory

36 – Cabarrus
Paul Newton (R)
+22% Trump
+24% Burr
+18% McCrory

37 – Mecklenburg
Jeff Jackson (D)
+60% Clinton
+54% Ross
+57% Cooper

38 – Mecklenburg
Joel Ford (D)
+54% Clinton
+46% Ross
+51% Cooper

39 – Mecklenburg
Dan Bishop (R)
+1% Trump
+12% Burr
+2% McCrory

40 – Mecklenburg
Joyce Waddell (D)
+54% Clinton
+47% Ross
+49% Cooper

41 – Mecklenburg
Jeff Tarte (R)
+4% Clinton
+4% Burr
+7% Cooper

42 – Alexander, Catawba
Andy Wells (R)
+43% Trump
+42% Burr
+35% McCrory

43 – Gaston
Kathy Harrington (R)
+31% Trump
+31% Burr
+26% McCrory

44 – Cleveland, Gaston, Lincoln
David Curtis (R)
+41% Trump
+40% Burr
+35% McCrory

45 – Iredell, Yadkin
OPEN SEAT
+42% Trump
+42% Burr
+33% McCrory

46 – Avery, Burke, Caldwell
Warren Daniel (R)
+47% Trump
+44% Burr
+37% McCrory

47 – Madison, McDowell, Mitchell, Polk, Rutherford, Yancey
Ralph Hise (R)
+44% Trump
+37% Burr
+30% McCrory

48 – Buncombe, Henderson, Transylvania
Chuck Edwards (R)
+28% Trump
+26% Burr
+17% McCrory

49 – Buncombe
Terry Van Duyn (D)
+25% Clinton
+24% Ross
+31% Cooper

50 – Cherokee, Clay, Graham, Haywood, Jackson, Macon, Swain
Jim Davis (R)
+35% Trump
+30% Burr
+19% McCrory

Previous Post Next Post

30 Comments

  • segmentation_fault August 11, 2017 at 10:48 pm

    Good job. Two qualms:

    19/21 in Cumberland/Hoke could still be struck down as a racial gerrymander. Your lines in Fayetteville are drawn based on where black precincts are.

    41 and the neighboring district in Mecklenburg could still be a racial gerrymander. A court would say by snaking the district along the Catawba River you are avoiding putting black voters into the district – a similar reason to which the current district was struck down.

    Keep in mind the map will have to pass muster with the Dem controlled state Supreme Court. That court will not cut them any slack.


    En Marche!

    • shamlet August 12, 2017 at 10:26 am

      19/21 can actually be cleaned up quite a bit with the new precincts in Cumberland. I don’t know if you’ve actually looked at it but Fayetteville is up there with parts of Alabama (Auburn-Opelika in particular) as the most frustrating place in the nation to draw in DRA because there are a bunch of absurdly large (15K-35K person) precincts. Splitting those (especially that D-leaning 15K precinct just south of downtown) will clean up the lines a bit without any real substantial issues.

      For Meck, I think you can pretty easily justify that as a political gerrymander. I think this map has a little bit of a problem in Meck, but it’s not with 41, it’s with the 37/40 line. I would keep 39 and 41 as is and then pie-slice 37-38-40 in the cleanest way possible.


      R, MD-7. Put not your trust in princes. Process is more important than outcome.

      • segmentation_fault August 12, 2017 at 1:21 pm

        Is the distinction between racial gerrymander and political gerrymander purely intent?


        En Marche!

  • rdelbov August 11, 2017 at 11:15 pm

    A very interesting 1st pass–It is very likely that the GOP could concede a Wake county seat –it could happen

    Your cumberland county seat is interesting. I thought about a Hoke county to suburban Cumberland county seat! I think the lines–that plunge can be smoothed out with very minor effect.

    I think I might move your Guilford county map around a bit.

    Only three seats in Guiford. One centered on Greensboro with the other GOP seat wrapping around it–Then the rest of the population in the Highpoint area attached to Randolph with the balance of that county going with Alamance.

    Mecklenburg county is a challenge. As long your stay out of Charlotte with seat #41- that is the Northern to southern seat then you might be okay

    Interesting stuff

  • rdelbov August 12, 2017 at 8:18 am

    On a second look and frankly this map deserves several looks I suspect that we see a more compact map. I like your Mecklenburg map and if the lines stay out of the city then perhaps it will stand-The lines for the whole county might need to be smoother. I personally would not concede a third senate seat in Wake county. Take the area of Dan Blue’s district out of the Wake-Franklin seat. Smooth the lines out and add one really good precinct to each of the other two GOP seats while removing a D one. I think that gives the GOP a chance at all three seats.

  • shamlet August 12, 2017 at 10:23 am

    Yup, this is literally exactly what I was thinking. The big question is if they try to tweak the county pods from what they called “optimal” which are decidedly not. In particular minor tweaks to 1 and 3 would put Bill Cook back in his seat and make 1 far safer – just move Beaufort to 1 and Hertford/Gates/Washington to 3.

    I am torn a bit on whether outright conceding a Wake seat (as they would have to do if they keep that Wake/Franklin pod) is the way to go. If you make a Wake/Durham pod you can get two seats that are at least Lean R but might be shaky given how the area is trending, and another seat of Franklin + Granville + Person + Caswell that’s Lean R but trending our way.


    R, MD-7. Put not your trust in princes. Process is more important than outcome.

    • rdelbov August 12, 2017 at 3:16 pm

      I think the GOP is set on the pods that show above. I think-they actually match(or I think they do) the common cause map of mods. I do like the idea of a Franklin-Granville-Person line of attack.

      I note that while I am focused on at 30 GOP seats we could steal the Pitt/Greene county seat in 2018. Don Davis is more liberal then that seat and it tight. Not taking race into account makes this seat a possible GOP pickup. Maybe Greg Murphy could win this seat in 2018?

      Focusing on 30 means that we can’t get too crazy. I would gamble/attempt to get 3 seats in Wake/Franklin. Mayve 2016 will be a high water mark til 2022? Clearly IMO it will be 3D-2R after that in Wake County and we might be happy to get it? The D proposal called for respecting city lines so the Mecklenburg lines could stand the test and frankly we still might lose that second seat but at 35 now we are playing with inside money. We might not hold a seat. Thinking about 35 if we try for 3 seats in Wake only get two so be it. Stuff happens. I think you can do a very clean map with giving the GOP a shot at 3 seats. The trick is that the Barefoot seat needs to be weakened from its current numbers by 4% or so. Give one or two good precincts to the current marginal R seats. It takes courage to do that! If not just do a map similar to one below and think that losing a marginal seat and gaining a lean R is better then losing two seats.

      • segmentation_fault August 12, 2017 at 4:16 pm

        I think Democrats will be putting up a stronger fight in legislative seats in 2018 than they have the past couple of cycles. With a couple of exceptions (like Tom Bradshaw running in SD-15 in 2014) Democrats have suffered from poor candidates. The filing deadline for 2016 being in December 2015 hurt because Democrats did not know they would be running against Trump or HB2. We had a random Native American rights activist running in suburban Wake SD-15 and a Bernie Bro running in South Charlotte against Dan Bishop.

        Having a governor helps with fundraising and the lack of a statewide race means that the side that is more energized will have stronger turnout. My two cents.


        En Marche!

        • rdelbov August 12, 2017 at 4:32 pm

          You could be right and I will fully admit that in 2014 the Hagan campaign did an outstanding job in turning out Ds in the state. There you have you it. I only complement a D campaign, no more then once a month, and the 12th I have already used up my quota. Yet I say look at 2014 turnout among Ds and compare it to 2010 and it was a big improvement. Naturally 2016 like 2012/2010/2014 was also a total all out fight for NC. In 2010,2012,2014 and 2016 national groups (on both sides) put in over 20 million dollars each into turnout efforts in NC. On the D side the turnout bucks showed in Presidential and Senate campaigns in those years.

          So what about 2018? I defer to my NC friends (and to RM) but to my way of thinking there is no way that the money for turnout efforts in NC will come anywhere close to 2010/2014 levels. So yes the Ds might have better candidates and even better funded legislative candidates in 2018 but the vast amount of resources that ginned up turnout on both sides from 2010 to 2016 was national money. I don’t see 20 million dollars coming into NC in 2018 for legislative races. What does all this mean?My guess is lower turnout in 2018 then 2014 or 2010. Will that help the GOP-historically it has but I guess we will see. Let’s be clear in the state senate we are talking (even under new maps) about 8 to 10 tight races. In the state house it could be 16 to 20 races. You might seen some 400K house races in some counties but still nothing like the spending associated with US senate races in 2014/2016.

          • rdelbov August 12, 2017 at 7:48 pm

            Just a note-apparently there was 2.1 million votes cast in 2006 while there was 3.1 million cast in 2014 with 4.6 million votes cast in NC in 2016

  • King Hushpuppy August 12, 2017 at 12:40 pm

    Thanks everyone for your comments. Some things:

    1. Based on my understanding of Stephenson, the clusters will not change (unless someone can find a more ‘optimal’ grouping). The way that mapmakers proceed is this: draw as many one county groupings as possible. Then, do the same for two county groupings. If your map has 12 2-county groupings but a map can be drawn with 13, you MUST use the second map. Proceed in this manner for 3-county groupings, 4 county-groupings, etc. I never really understood the county grouping requirement, but the ‘optimum’ county clusters are exactly what would be obtained if done in the manner just described.

    The only way, I think, to get around the Stephenson optimum county grouping requirement is to draw VRA districts first. Then the ‘optimum map’ would look different based on how those are drawn. Legislators have said this time they will not consider race at all, so they’re not going to do this.

    So these are almost certainly the clusters that will be used. Could they, at the last minute, pull a surprise and say they’ve found a county grouping that works even better? Maybe, but I doubt they’re in the mood for stunts.

    2. Rdelbov, I actually tried what you suggested in the Guilford cluster and it works better for the GOP. The problem is the Whole Counties Provision – better to split one county in a cluster multiple times than to split two counties. If they draw something like that, plaintiffs can get the map thrown out because they split a county when it wasn’t necessary.

    3. Based on the clusters, Durham goes with Person and Granville (wasting a bunch of Republican votes) and Wake is paired with Franklin. Three GOP seats can theoretically be drawn in the Wake-Franklin cluster. But it is very, very risky. After drawing the 2 Democratic sinks, the remaining area to be divided between the GOP seats is a 5% Burr win but also a 5% McCrory loss. McCrory’s performance was pretty much the worst for any Republican in Wake County in recent history, but about what I think the GOP would perform in a blue wave. Therefore, drawing such a map would make it very possible for Democrats to take 5/5 seats in the area. In my view, it’s just too risky.

    Conceding a seat to the Democrats in Wake allows the creation of two seats there that went for Trump and McCrory. If someone in suburban Wake County voted for McCrory despite HB 2, you’d have to think their chances of supporting any Democrat are practically nil. I definitely think it’s best to play it safe.

    4. On the other hand, I would not play it safe in Mecklenburg. One could concede 41 off the bat and avoid drawing a non-compact district, but I think that would be a mistake. I think there are enough voters who support Republicans at the local level to draw two GOP-leaning districts. And last year’s results bear this out: the Republican in 41 won reelection by 14% and the Republican in 39 won by 13% (despite being the architect of HB 2!).

    Thanks everyone for your comments and suggestions. I might try a House map, but that will take longer obviously.

    • Jon August 12, 2017 at 11:42 pm

      There tends to be one important exception to pod systems above for states with strict county integrity requirements: If doing this ends up causing a split of a county that is smaller than ideal population for a district, but it’s possible to have drawn a map avoiding that (by say locking in a four county configuration before a three), they’ll need to do that.
      And there are some states in which having two districts cross a county line is just as illegal as having a county below ideal population of a district being split. (e.g. MO where a court drawn senate map that had both of Springfield’s districts crossing the county line in the case of a county too big for one senate district but two small for two senate districts was ruled by the state supreme court to be illegal, and required to be redrawn in a way that kept one of the two districts entirely within the county.)


      45, M, MO-02

      • rdelbov August 13, 2017 at 12:44 pm

        Exactly and in 2020 the GOP might actually use this pod system to keep 60% control in Senate. It will be hard to keep a third R seat in Wale county but if you use Franlin-Granville-Person as the foundation for a seat that offsets a lost seat in Wake. In Mecklendburg there looks like it be also 5.2 or so seats you might end up attaching Irendell or another smaller county to North part of the county for a GOP seat

    • rdelbov August 14, 2017 at 1:12 pm

      Just a thought on the state house map–IMO Wake, Mecklenburg and Cumberland counties are the only ones that really matter. In Forsyth and Guiford county seats the lines can easily be smoothed out while making any partisand changes. Ditto for rural seats. The seat GOP seat in Cumberland will be interesting while in Wake and Mecklenburg counties I believe the GOP will need to add to their numbers by conceding marginal seats to the Ds. The Ds have won 5 marginal house seats in those two counties and rather trying to save all five its time to concede a couple!!

  • shamlet August 15, 2017 at 9:49 am

    Ok, this makes more sense now: Cook is likely to retire and Steinburg is probably going to run for the open seat. https://outerbanksvoice.com/2017/08/14/district-remapping-would-open-door-for-a-new-state-senator/


    R, MD-7. Put not your trust in princes. Process is more important than outcome.

    • segmentation_fault August 15, 2017 at 3:46 pm

      They made Steinburg’s house seat a lot swingier by dropping Currituck and adding Washington & Bertie. Probably an Obama-Trump district though.


      En Marche!

  • rdelbov August 15, 2017 at 12:18 pm

    I would be surprised if we do not see 4 or 5 senators either retire or change seats. Perhaps the same or more due to house changes.

    I suspect the number of changed seats will not be all that more then normal for any house or senate elections.

  • shamlet August 15, 2017 at 5:45 pm

    KingHushpuppy (or anyone else) – I have a drf ready of what I think the House plan will be but I won’t have time to write it up for a while. If anyone is interested email the site and I will send it to you.


    R, MD-7. Put not your trust in princes. Process is more important than outcome.

  • rdelbov August 15, 2017 at 6:16 pm

    I look forward to looking at it

  • segmentation_fault August 16, 2017 at 3:53 pm

    @Shamlet

    I’m trying it out myself for the House map. I noticed that for the Lee/Harnett/Johnston/Wayne/Greene/Sampson/Bladen county group the deviation is very close to the legal maximum, i.e. all the districts will be very overpopulated. Is this what you have?

    (Also, that means you can’t draw a legal version without serious precinct splitting)


    En Marche!

    • shamlet August 16, 2017 at 4:06 pm

      Yep. All those seats are like 10 people under the 83435 maximum. So I just left a couple overpopulated seats.


      R, MD-7. Put not your trust in princes. Process is more important than outcome.

      • segmentation_fault August 16, 2017 at 4:37 pm

        Seems like it hurts Republicans to do that although if they use the methodology for forming county groups that King Hushpuppy described above I guess they didn’t have a choice.


        En Marche!

    • Jon August 16, 2017 at 6:55 pm

      This is a case in which if adding one more county to the group would result in a cluster significantly closer to an ideal multiple of seats it would have been preferred by me.

      In fact, I’m of the opinion that while drawing first the 1 county clusters ; then 2 county around them; then 3 around previous etc. is a good first approximation, the map then should almost always have their clusters tweaked to an overall more natural settings.
      (Alternately, instead of considering .95 – 1.05 , 1.9 – 2.1 , 2.85 – 3.15 , 3.8 – 4.2 … all valid clusters only consider … 2.9 – 3.1 , 3.9 – 4.1 … valid ones if its two or more counties.)


      45, M, MO-02

      • shamlet August 16, 2017 at 7:01 pm

        I would do it by making the metric the average number of counties per cluster rather than the most clusters with the fewest number of counties. It’s absurd to think a 10-county and a 3-county cluster is always better than a 4-county and a 9-county one. That still means you have to draw as many 1-county pods as possible and likely forces you to maximize 2-county ones too, but beyond that you have a bit more freedom to make things cleaner.


        R, MD-7. Put not your trust in princes. Process is more important than outcome.

  • shamlet August 20, 2017 at 9:27 pm

    Major kudos on getting it basically exactly right!


    R, MD-7. Put not your trust in princes. Process is more important than outcome.

    • rdelbov August 21, 2017 at 10:35 am

      The major divergence was in the Cumberland county arrangement–I like his map better but that is another matter.

    • King Hushpuppy August 21, 2017 at 3:29 pm

      I got the Hoke/Cumberland idea from your map, actually. I think limiting 21 to Fayetteville is actually better for Meredith, but they wanted to keep Ben Clark in the Senate.

  • rdelbov August 21, 2017 at 3:47 pm

    I think the Ds will complain the most about Guiford and maybe Fayetteville area. They might complain about Mecklenburg as well.
    I personally think the judges will let this map stand.

    As you noted you did a great job on your map.

  • rdelbov August 21, 2017 at 4:19 pm

    Political data is out!!

    http://www.heraldsun.com/news/state/article168382892.html

  • Leave a Reply

    Do NOT follow this link or you will be banned from the site!