North Carolina 2020s Senate Redistricting

This is the North Carolina State Senate edition of my “2020s Redistricting” series. Data in DRA has been modified to reflect 2020 population projections. You can see my proposed 2020s congressional map HERE.

Drawing maps for the North Carolina General Assembly is challenging (and also restrictive) due to the Whole Counties provision of the state constitution. Legislators in the past also had to draw districts to conform to the Voting Rights Act. Given recent court decisions, I have decided to draw a map which takes race out of the equation completely. This, of course, might lead to other constitutional challenges (see SD-1 below).

With that in mind, I have also drawn the map the way Republican lawmakers might do so should they be in control of the General Assembly in 2021 (and assuming partisan gerrymandering is still allowed). Despite surging growth in Democratic urban counties and stagnation in Republican rural ones, the result is a map only slightly more favorable to Team Blue. The districts are reviewed below (partisan results once again reflect the two-party vote).

First, the whole map:

1 – Bertie, Camden, Chowan, Currituck, Dare, Gates, Hyde, Pamlico, Pasquotank, Perquimans, Tyrrell, Washington
Norm Sanderson, R-Pasquotank
52.0% Romney
57.6% Trump
The 1st loses Beaufort County and adds Pamlico, Washington, Tyrrell, and Bertie counties. This last one is particularly controversial because it has the highest percentage of black residents in the state. Sanderson (should he still be in the State Senate) might be vulnerable in a primary, because almost all of the district except for his home county of Pamlico is new to him. This is an ancestrally very Democratic district but should be likely GOP now.

2 – Beaufort, Carteret, Craven
Bill Cook, R-Beaufort
63.0% Romney
65.4% Trump
The 2nd loses exchanges Pamlico County for Beaufort County. It’s still a safe GOP seat.

3 – Granville, Halifax, Hertford, Northampton, Vance, Warren
Erica Smith-Ingram, D-Northampton
36.6% Romney
39.8% Trump
The 3rd is no longer majority black, but is still safe Democratic and should keep electing a black senator.

4 – Edgecombe, Franklin, Nash
Angela Davis, D-Nash
45.6% Romney
47.8% Trump
Unfortunately, the Whole County Provision pretty much requires drawing something like this that wastes a bunch of GOP votes. The saving grace for Republicans might be the VRA – this district includes majority black Edgecombe County in a district that’s majority white.

5 – Martin, Pitt
Don Davis, D-Pitt
46.6% Romney
46.6% Trump
The Whole County Provision also requires combining light blue Pitt with another county, in this case Martin. Again – a whole lot of wasted GOP votes.

6 – Duplin, Greene, Jones, Lenoir, Onslow (part)
Harry Brown, R-Onslow
56.1% Romney
59.4% Trump
The current incumbent (and a good possibility to succeed Phil Berger as Senate pro tem) lives just outside this district, but that can be easily corrected. This is inelastic GOP territory which is practically wave-proof.

7 – Wayne, Wilson
Louis Pate, R-Wayne AND Rick Horner, R-Wilson
50.8% Romney
52.1% Trump
This is a marginally GOP district at the presidential level where unfortunately white flight is a real problem. The competition here will be fierce. Obviously, Democrats will need strong turnout from black voters to win here.

8 – Brunswick (part), New Hanover (part)
Bill Rabon, R-Brunswick
57.0% Romney
56.5% Trump
This is an educated, fairly affluent seat on the coast where Trump 2016 slightly underperformed Romney 2012. Democrats will still need a wave or a scandal to win here.

9 – Onslow (part), New Hanover (part), Pender (part)
Michael Lee, R-New Hanover
54.7% Romney
59.2% Trump
This seat trended rightward last year and should continue to do so in the future as it burgeons with coastal conservative retirees. Not safe GOP territory, but getting there.

10 – Bladen, Cumberland (part), Sampson
Brent Jackson, R-Sampson AND Wesley Meredith, R-Cumberland
55.1% Romney
57.4% Trump
District 10 shifts to the west and becomes a seat based on rural areas near Fayetteville. This is a fairly inelastic GOP seat and the chances of it returning to its historically Democratic roots are very questionable.

11 – Johnston
Open seat
64.0% Romney
65.7% Trump
Another seat where clever GOP gerrymandering is thwarted by the Whole Counties Provision. A lot of GOP votes are wasted in this 65.7% Trump seat, which is unfortunate because the Triangle and Eastern NC could really use some reinforcement from Johnston County. Best case scenario for the GOP? The county grows much faster than expected, forcing it outside of the 5% constitutionally permitted population deviation.

12 – Harnett, Wake (part)
Tamara Barringer, R-Wake AND Ronald Rabin, R-Harnett
58.6% Romney
57.5% Trump
District 12 loses Lee and instead takes in a number of precincts in Southern Wake County. The Wake portion of this district voted for Trump but this is an area that could get bluer with time.

13 – Brunswick (part), Columbus, Robeson (part)
Danny Earl Britt, R-Robeson
54.0% Romney
60.1% Trump
This district zoomed to the right in 2016 and shouldn’t be a difficult hold for Republicans.

14 – Wake (part)
Dan Blue, D-Wake
22.2% Romney
21.4% Trump
This East Raleigh district is plurality black and should continue to elect black Democrats to the upper house.

15 – Wake (part)
Chad Barefoot, R-Wake
57.6% Romney
54.7% Trump
This is about as safe a Trump district one can draw entirely in Wake County, and will probably get less and less safe with time. In this and other educated, suburban districts, the Romney number is probably only attainable in a GOP wave year.

16 – Wake (part)
Jay Chaudhuri, D-Wake
34.3% Romney
28.6% Trump
This vote sink contains Morrisville, Cary, and Raleigh and has many white liberals who work in high tech. Obviously, it’s safe Democratic.

17 – Wake (part)
Open seat
49.0% Romney
41.4% Trump
Mitt Romney was pretty much a perfect fit for this area in 2012, and voters here still rejected him for four more years of Obama. Then, in 2016 the GOP’s numbers further tanked with Trump heading the ticket. Republicans will need a wave, and even that might not be enough.

18 – Wake (part)
John Alexander, R-Wake
48.8% Romney
41.1% Trump
See the description for the previous district.

19 – Hoke, Lee, Moore
Ben Clark, D-Cumberland
57.0% Romney
58.7% Trump
This combination of counties in the central part of the state makes for a safe GOP seat. The incumbent Ben Clark is black and might be the most conservative Democrat in the Senate. This won’t be nearly enough to make him palatable to deep red Moore County, which makes up almost half of the district.

20 – Durham (part)
Floyd McKissick, D-Durham
14.2% Romney
10.3% Trump
A district centered on downtown Durham, it’s only plurality black and the most pro-Clinton seat in the state. Trump almost sunk into single digits here. Obviously, it’s more than safe for any black Democrat.

21 – Cumberland (part)
Open seat
27.6% Romney
28.9% Trump
Another diverse district, this one focused on the Fayetteville and Fort Bragg area. Solidly Democratic, obviously.

22 – Chatham, Durham (part)
Mike Woodard, D-Durham
40.5% Romney
35.7% Trump
This district wraps around the 20th and takes in the remainder of Durham as well as light blue Chatham County. It’s safe Democratic and is trending away from the GOP.

23 – Caswell, Orange, Person
Valerie Foushee, D-Orange
35.7% Romney
33.0% Trump
Despite containing three counties (two of them rural and conservative) this district is absolutely dominated by the politics of ultra-liberal Chapel Hill and should continue to elect a progressive Democrat.

24 – Alamance, Guilford (part)
Rick Gunn, R-Alamance
57.7% Romney
57.1% Trump
This district trades in part of Randolph County for part of Guilford and becomes significantly bluer, but should still be a fairly easy GOP hold. It could still fall in a wave, though.

25 – Anson, Richmond, Robeson (part), Scotland, Union (part)
Tom McInnis, R-Richmond
48.6% Romney
56.5% Trump
An Obama district that Democrats will probably have to take back to win the majority under this map. Given current trends, that might be tough – especially with a GOP incumbent.

26 – Mecklenburg (part)
Open seat
46.3% Romney
39.6% Trump
Phil Berger’s seat in the Triad gets transported to the Charlotte suburbs, and it’s almost certainly a Democratic pickup.

27 – Guilford (part)
Trudy Wade, R-Guilford
59.8% Romney
54.6% Trump
This seat in suburban Guilford used to be safe GOP, but is more competitive in the age of Trump. Still a seat that favors Republicans, but perhaps not wave-proof.

28 – Guilford (part)
Gladys Robinson, D-Guilford
20.1% Romney
19.5% Trump
The (only plurality black) Guilford vote sink, which attempts to pick up as many Democratic precincts in Greensboro as possible. A black Democrat would be strongly favored in a primary here.

29 – Guilford (part), Montgomery, Randolph
Jerry Tillman, R-Randolph
62.8% Romney
67.3% Trump
Despite taking in much of High Point (which could raise potential VRA issues), this district is more than safe for any Republican.

30 – Rockingham, Stokes, Surry
Phil Berger, R-Rockingham
65.7% Romney
71.9% Trump
Phil Berger’s home county of Rockingham loses its portion of Guilford County and trades it for much more Republican Surry and Stokes counties, making it a solid GOP seat.

31 – Forsyth (part), Yadkin
Joyce Krawiec, R-Forsyth
65.6% Romney
64.2% Trump
A seat based in the Winston-Salem suburbs which changes only marginally from its predecessor. It’s safe territory for any Republican.

32 – Forsyth (part)
Paul Lowe, D-Forsyth
28.0% Romney
26.8% Trump
This seat is plurality white now, but it’s still a safe Democratic seat.

33 – Davidson, Davie
Andrew Brock, R-Davie AND Cathy Dunn, R-Davidson
70.9% Romney
74.9% Trump
Combining Davidson and Davie counties creates the most pro-Trump seat in the state. Safe GOP territory, obviously.

34 – Rowan, Stanly
Open seat
65.4% Romney
71.0% Trump
Another safe GOP seat in the Piedmont.

35 – Union (part)
Tommy Tucker, R-Union
63.5% Romney
63.8% Trump
Union County is unlikely to become swing territory any time soon, so this should be another safe GOP seat.

36 – Cabarrus
Paul Newton, R-Cabarrus
60.1% Romney
60.2% Trump
Under current population projections, Cabarrus County falls just within the 5% deviation subjecting it to the Whole County Provision, so it gets its own district here. Safe GOP, but this seat is gradually getting bluer.

37 – Mecklenburg (part)
Open seat
21.5% Romney
21.2% Trump
A black plurality seat in West Charlotte, safe Democratic.

38 – Mecklenburg (part)
Joyce Waddell, D-Mecklenburg
19.1% Romney
18.3% Trump
North Charlotte, plurality black, safe Democratic as well.

39 – Mecklenburg (part)
Dan Bishop, R-Mecklenburg
62.0% Romney
52.9% Trump
Look at those partisan numbers – Trump really tanked here. Unfortunately, this is about as safe a GOP seat as one can draw entirely in Mecklenburg County, and it’s a swing seat.

40 – Mecklenburg (part)
Open seat
23.9% Romney
21.8% Trump
A western Charlotte seat that is the most racially diverse in the entire state. Obviously, the general election winner will be determined in the Democratic primary.

41 – Lincoln, Mecklenburg (part)
David Curtis, R-Lincoln AND Joel Ford, D-Mecklenburg AND Jeff Tarte, R-Mecklenburg
62.7% Romney
61.5% Trump
North Mecklenburg Republicans get some welcome assistance from ultra-conservative rural Lincoln County. For now, it’s safe GOP territory.

42 – Caldwell (part), Catawba
Andy Wells, R-Catawba
66.8% Romney
72.0% Trump
Solid GOP.

43 – Gaston (part)
Kathy Harrington, R-Gaston
61.8% Romney
65.0% Trump
And another one …

44 – Alexander, Iredell
Open seat
66.7% Romney
70.7% Trump
And yet another safe GOP seat. This one is open. In a primary, an Iredell County Republican would be heavily favored.

45 – Alleghany, Ashe, Caldwell (part), Watauga, Wilkes
Deanna Ballard, R-Watauga AND Shirley Randleman, R-Wilkes
62.8% Romney
67.0% Trump
Safe GOP again. (Have you noticed that Republican votes are increasingly packed in this part of the state?)

46 – Burke, Cleveland, Gaston (part)
Warren Daniel, R-Burke
62.2% Romney
68.8% Trump
Again, not much to say here. Safe GOP.

47 – Avery, Madison, McDowell, Mitchell, Polk, Rutherford, Yancey
Ralph Hise, R-Mitchell
65.2% Romney
72.4% Trump
Adds Avery County. Otherwise, no changes. Safe GOP.

48 – Buncombe (part), Henderson, Transylvania
Chuck Edwards, R-Henderson
62.2% Romney
63.8% Trump
Exchanges some bluer precincts in Buncombe County with the most red ones. Other than that, no changes and a fairly safe GOP seat.

49 – Buncombe (part)
Terry Van Duyn, D-Buncombe
38.7% Romney
36.8% Trump
The Buncombe vote sink, and the only Clinton seat west of Charlotte. Sheds most of its northern precincts to District 48.

50 – Cherokee, Clay, Graham, Haywood, Jackson, Macon, Swain
Jim Davis, R-Cherokee
61.0% Romney
67.6% Trump
Unlike every other district, the 50th doesn’t change at all. At nearly 68% Trump, it’s hard to see any Democrat having a shot here.

The breakdown: 31 Romney seats, 32 Trump seats. Democrats need to hold all the Clinton seats and win 3 Trump seats to roll back the GOP’s supermajorities; they need to hold all the Clinton seats and win 7 Trump seats to break even, win 8 to take a majority.

I would welcome any feedback, especially when it comes to identifying potential VRA issues. Unlike my congressional map, I’m not at all confident that this is the strongest map for Republicans, especially in the eastern part of the state. I’ll draw another map incorporating any recommendations I receive here, so let me know if you have comments or any questions.

Previous Post Next Post


  • GOPTarHeel May 30, 2017 at 10:51 am

    What are you basing the population numbers off of? Last time I check the county population projections, Mecklenburg wouldn’t need any more seats to to be attached to another county-remember its five current seats are dramatically underpopulated. Wake would be at 5 seats as well.

    Once I get an answer on that I have some questions about your choices for the county pods.

    R/NC-13. I'll never regret a vote that resulted in Neil Gorsuch.

    • King Hushpuppy May 30, 2017 at 3:09 pm

      I calculated the population numbers by taking the 2016 population estimates (Census Bureau) and percentage growth rate from 2015 to 2016 and extrapolating that forward to 2020. For example, if a county grew 1.5% last year, I would apply that same rate to this year, and the next year, etc.

      The Office of State Budget and Management has population projections on their site. Theirs differs slightly from mine, but they also require Wake and Mecklenburg to be paired with another county: I’m not sure if the Census Bureau keeps numbers of their own. Do you have a view as to which is likely to be most accurate?

      • GOPTarHeel May 30, 2017 at 3:51 pm

        Guess it’s been awhile since I’ve looked at the OSBM Wake and Meck numbers, since you’re right. Those are insane growth numbers. I’ll give this some more thought. I do question the utility of drawing insane-looking Likely D seats in the Charlotte and Raleigh burbs. Seems to me like any marginal chance of winning them is outweighed by the “ick” factor that judges seem to have when they see maps like this. I’m also very wary of putting Bertie in a GOP seat and in the way you’ve cracked some of the SENC counties. Thanks for your hard work calculating the Trump numbers.

        R/NC-13. I'll never regret a vote that resulted in Neil Gorsuch.

        • King Hushpuppy May 30, 2017 at 5:04 pm

          Thanks. I definitely agree the map can be made cleaner in some parts. The current Senate map is pretty clean outside of the VRA districts. I’m not sure what to do about Bertie County as I’m not clear on what the current state of the law is regarding the VRA. For the last redistricting, legislators drew the VRA districts first and then followed the Whole County Provision. Do you think they will continue with this practice?

          • GOPTarHeel May 30, 2017 at 5:15 pm

            For the Senate map, probably not. If they’re willing to accept BVAP 40%+ seats instead of BVAP majority they won’t need to do anything to avoid reducing African American representation. But they’ll try and avoid any kind of vote dilution claim they could be triggered by splitting the Northeast up between GOP districts. I think they’ll calculate the absolute minimum number of county splits mathematically and then work from there.

            R/NC-13. I'll never regret a vote that resulted in Neil Gorsuch.

            • Jon May 30, 2017 at 7:23 pm

              Under my understanding of the recent SCOTUS decision all mechanical racial threshold percentages are highly suspect.

              I’d recommend first draft not looking at racial data at all but simply complying with NC whole county provisions.
              Then check on the case by case basis, but the Circuit has just basically said that for the territory covered by NC-01 that Whites don’t sufficiently block vote to defeat AAs and so you only violate VRA is you do consider race.
              I think this will pretty much mean that all the districts composed solely of whole counties will be okay.

              Now for those counties split between exactly two districts, it would probably be a bad idea to have the line splitting an AA community down the middle; have that community all in one district or the other.

              45, M, MO-02

              • rdelbov May 30, 2017 at 7:34 pm

                I agree-check for deviations in population and the county split rules. Since race cannot be a factor you need to avoid splitting a county two times. Better to be ugly then split a county twice or three times.

  • Jon May 30, 2017 at 7:11 pm

    What was max deviation for the current 2011 senate maps?
    What is your estimated max deviation for your map?

    For a deviation likely to be chosen for state legislative maps, there’s still too many counties in which part of a county is listed multiple times in a district that crosses county lines. (District numbers 6, 8, 9, 13, 24, 25, and 29).

    Should the state choose +/5% there would be expected to be between zero and two such cases.

    At +/- 3% I’d still expect at most five such cases.

    Particularly suspect districts are #8 & #9 composed entirely of split counties with one county shared between them; by shuffling around 6, 8, 9, 13, and 25, it ought to be possible to remove at least one of those splits.

    45, M, MO-02

    • GOPTarHeel May 30, 2017 at 11:08 pm

      They used 5%

      R/NC-13. I'll never regret a vote that resulted in Neil Gorsuch.

      • Jon May 30, 2017 at 11:24 pm

        Yup; poster would need to spend a few hours shuffling the partial counties around to clean it up to whole counties while staying within +/- 5% then. There may be a way or two to not split any county smaller than an ideal senate seat. But even if there isn’t one, then unless counties in corners have very awkward populations then there will be a map possible with only one county that’s smaller than an ideal senate seat being split.

        45, M, MO-02

  • rdelbov June 1, 2017 at 5:49 pm

    Re NC counties–Mecklenburg and Wake should be at 5 senate seats-assuming 5% deviation. My Wake county guess would be 3D seats and 2 R seats. Not sure Mecklenburg can be done that way but I like some of these maps. My guess would be trying to limit D seats to 17 or 18 seats. Fairly similar to current map

    • Jon June 1, 2017 at 6:56 pm

      Small nitpicking note that might or might not apply:
      While it’s standard in many states with county integrity requirements for counties whose House Seats is at or above 4.9 ideal population for house seats to simply round the county to the nearest whole number, it’s not quite the case with Senate seats. Unless the county is very close (4.9 to 5.1) then states with whole county requirements usually have the last seat crossing the boundary, preferably to an adjoining county that when added together brings the two counties combined to very close to a multiple of ideal population for senate seats.
      TN: Shelby County : Even when it had ideal population for 5.2 Senate seats, it still had a small part of the sixth senate seat rather than being whole numbered down to 5. Keeping that small fraction made complying with the state’s requirement that no county smaller than the ideal population of a senate seat be split easier.

      45, M, MO-02

  • rdelbov August 1, 2017 at 10:39 am

    We will be seeing a redraw of the NC senate map this month. Not sure we can save both Wake county seats–Mecklenburg as well. I see an easy fix in Guiford county but stay tuned.

    • rdelbov August 1, 2017 at 10:43 am

      I know the GOP does not want a huge redo but my solution to the Cumberland county seat is to attach Hoke county to Moore/Randolph county seat. That will move voters around in that area. I know the GOP would prefer a minimum redo but that may not be possible.

    • Jon August 1, 2017 at 7:21 pm

      Not use this proposed map is the best starting place (even if it didn’t have the extra county splits) for 2017; they are going to have to use 2010 census numbers instead of 2020 projections or even 2015/6 estimates.

      45, M, MO-02

    Leave a Reply

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