Colorado 2020 U.S. House Redistricting Discussion

[Oryxslayer]

First off, Colorado now has a commission. The structure of Colorado's commission is very similar to California's in that there are a large number of commissioners and any map needs 2/3s of them to approve a plan. Now, Colorado's commission is carefully worded to specify party members as the criteria for being selected  as a partisan/indie commissioner. Since we all know that Colorado's Indie voters consistently break blue (Boulder has plurality Indie registered for instance) there is the potential that the commission ends up slanted and just selects D-favoring maps. However, lets ignore this potential for now and move  on to the main event: Denver.

If the trifecta had their way with the maps, Denver would be carved six ways to Sunday and the entire metro region would become Safe for their blue incumbents. However, with commission in place, there are really just two general plans in regard to the city.

The first general plan is keeping most or all of the city inside one district. This is obviously a democratic pack, but it keeps the the city COI intact. If such a district is drawn, the 8th district is almost guaranteed to be based out of Douglas County. This means the seat is going to be both more white and more GOP than the other seats in the region. This style  of map would focus on preserving city, town, and county COI's.



The second option has the side effect of splitting Denver, which is why the democrats may have their interest groups lobbying the commission hard for this type of plan. This general option is a minority district. The district could resemble the  lines below, or it could drop parts of Adams in favor of Aurora. This type of plan favors minority COIs over city ones. Such a plan would release the white parts  of Denver to her suburbs. Even if this chunk of the city is kept whole, it still will push the suburban seats towards the democrats because of it's hyper-democratic partisanship. Such a plan does not guarantee a seat based out of Douglas, nor does it guarantee such a seat even favors the Republicans.

[Oryxslayer]

https://davesredistricting.org/join/0e67ae47-147a-43bf-af1a-112973325418
4D-2C-2R map
Gives both sides something they'd want.  Majority-minority Denver seat and 2 other suburban seats have areas of Denver that make them safe D.  Boulder+Fort Collins seat safe D.  CO Springs and Eastern CO seats are safe R, and competitive seats in western CO and the southern suburbs of Denver.

Firstly, given it's an independent commission it doesn't matter what both sides want. Secondly, given Colorado has a Democratic trifecta, why would the Republicans ever get anything they want in redistricting? Thirdly, your "competitive" seats are R+7 (6th) and R+9 (5th) so really you've just drawn a 4-4 map, or a light Republican gerrymander.

It's a bipartisan commission. 

It's as much of a bipartisan commission as Arizona has a "bipartisan commission"

Colorado's commission is a CA style commission. 12 Commissioners, 4  D, 4 R, 4 Indie. Any map needs 2/3s approval, and at least 2 indies need to be in that 2/3s. now, like I mentioned above, the indies could be stacked. However, the more commissioners there are, the lower chance  of this happening. AZ for instance has a simple majority to adopt their maps with a 5 member commission, putting all power in that single indie.

[Oryxslayer]

https://davesredistricting.org/join/0e67ae47-147a-43bf-af1a-112973325418
4D-2C-2R map
Gives both sides something they'd want.  Majority-minority Denver seat and 2 other suburban seats have areas of Denver that make them safe D.  Boulder+Fort Collins seat safe D.  CO Springs and Eastern CO seats are safe R, and competitive seats in western CO and the southern suburbs of Denver.

Firstly, given it's an independent commission it doesn't matter what both sides want. Secondly, given Colorado has a Democratic trifecta, why would the Republicans ever get anything they want in redistricting? Thirdly, your "competitive" seats are R+7 (6th) and R+9 (5th) so really you've just drawn a 4-4 map, or a light Republican gerrymander.

It's a bipartisan commission. 

It's as much of a bipartisan commission as Arizona has a "bipartisan commission"

Colorado's commission is a CA style commission. 12 Commissioners, 4  D, 4 R, 4 Indie. Any map needs 2/3s approval, and at least 2 indies need to be in that 2/3s. now, like I mentioned above, the indies could be stacked. However, the more commissioners there are, the lower chance  of this happening. AZ for instance has a simple majority to adopt their maps with a 5 member commission, putting all power in that single indie.

This is particularly likely to happen in Western states with a lot of "indies" who all lean one way, to the point where I would go in with the assumption of a soft Dem map in CO and AZ and a soft GOP map if e.g. the OK/NE commission initiatives succeed.  In, say, Michigan, I would be more confident that the independents on the commission have unique views not aligned with either party. 

Also some southern states with a lot of Republican voters who are still registered dem could have this issue if they were to get bipartisan commissions.  Like KY.

Very well could happen in Arkansas if their commission actually comes into existence. Every Dem would need to be a AA for the Dems to have any chance at pushing for an AA seat on the Congressional level. If not, then the Dems would probably be encouraged to trade GOP authority over the Congressional map for a greater amount of Dem seats on the State legislative maps. Oklahoma is also pushing for a commission, but if it successfully becomes law, Dems don't really have much to worry about on the Congressional level. Their only possible desire, keeping OKC whole, is fairly common sense and would occur under any fair map.

[Oryxslayer]

And CO isn't required to do so. Whites in Colorado aren't uniformly opposed to electing a Hispanic candidate, so it would just be pure packing of minorities to dilute Democratic voting strength to create a Hispanic VRA seat there.

It doesnt really dilute Democratic voting strength, unless you're making tentacle districts out of Denver

If anything, Dems want the minority seat in this situation. It's being built out of an area that is already very blue, so no dems are getting cut. In exchange for slight packing, Dems get to seriously cut and spread out Denver, something that would normally only be possible under D-gerry rules.

[Oryxslayer]

I haven't been following predictions about their lines--what's the likelihood that Boebert's district gets more friendly to Democrats?

Depends. The big question that will be unveiled tomorrow is the fate of the West Slope. All the assorted counties west or part of the Rockies are very close to 1 district, so pairing them together seems like the most natural community ever. However, this forces a Pueblo + Weld linkage via the rurals. While this makes sense geographically, it does force two disparate cities together. The alternative is keeping Pueblo in 3, which would mean another cut into the western side of the state in the north. There's inertia keeping Pueblo with the 3rd, but there was advocacy in favor of a western seat

I think Boebert would prefer things partisan-wise if Pueblo was kept in the seat. Dropping Pueblo does give her some more GOP rural turf, but it also unites all the ski towns. There are more potential D voters there than in pueblo.

[Oryxslayer]

Preemptive reminder that the map released today, and the legislative map released in a few days, are not final. Aside from the obvious that this is drawn using the ACS data - requiring minor adjustments later - these maps are meant to simply serve as a launching point for discussion. The months between now and the census delivery date intend to be used for public forums and discussion. The commission will then take this input and final census numbers and release a final product in the fall. So, the main things to look for are not the minute cuts and pairings, but rather the overarching vision which is unlikely to change. Districts will see adjustments, but their general location, orientation, and the purpose behind pairing communities to create said constructs are unlikely to change.

[Oryxslayer]

Okay, so Pueblo-Weld is in. No Denver minority access seat, unless Aurora has enough on its own under the new numbers.

Functionally, I'm not sure what it the up or downsides just from glancing. 6 gets much safer for Ds. 7 is the new seat, 8 replaces the old 7. New 7 is probably more Dem than most hypthetical outcomes that kept the emphasis on Douglass. I suspect the Adams-Weld weirdness will get cleaned up through citizen input.

I'm not quite sure about the new 3. Gets more blue ski turf, but also the red counties to the east of the Rockies but west of Colorado Springs.

It seems like their guiding light is keeping the cross-county cities like Westminster and Aurora intact.

[Oryxslayer]

Looks like it was a 4-4 map based on 2018 data.

So 5-3 based on 2020 data. That race was much closer than the Pres or even the senate election.

[Oryxslayer]

Apparently the new 8 (old 7) is an attempt to get a Hispanic Rep using Greeley and Adams to get a seat >30% Hispanic under the current data. Honestly, may be a better option than the Aurora-Denver minority seat some proposed, but probably less effective than the Adam-Denver one.

https://twitter.com/Nickyg113/status/1407796458275934213?s=20

Boeberts seat got bluer.

[Oryxslayer]

https://twitter.com/ECaliberSeven/status/1407799392153833472?s=20

There seems to be some confusion over the Boebert numbers in 3, which is what I expected when the map was revealed. Takeaway should be 7 is around +9 Biden. Hard to see that electing a Republican.

[Oryxslayer]

Some rough GIS work on my part puts 3 around 52.2% Trump, 45.4% Biden. Arguably the biggest effect is putting a bunch of Boebert's Pueblo-based challengers in a different seat.

[Oryxslayer]

Given trends the GOP might have to worry more about 5 than 3 later on. 5 is Dem trending but might not be competitive until 2028.

Agreed - CO-5 could easily be competitive by 2026 or 2028,  that was pretty much inevitable in any map.

The fifth was always going to be just Colorado Springs and her environs, so the seat becomes competitive under any map the moment D's make El Paso county competitive.

[Oryxslayer]

This is an interesting map. I'm not sure why they gave part of El Paso County to Buck's district, rather than keeping the whole county intact as its own congressional district. But otherwise, the map makes sense, and pretty closely correlates with what I expected. This would be a 5 D-3 R map in most years, assuming that Republicans don't reverse the Democratic rise in Douglas County.

The El Paso cut is only done for Pop reasons, CD5 remains nested in the county. DRA's 2019 estimates have the county just on the edge of 1 cd, but given growth it likely surpassed its allotment in their internal data. It and the rest of the front range is also likely to benefit from the final census numbers, at the expense of the south of the state and Pueblo region, given the expectation of a rural Hispanic undercount.

[Oryxslayer]

The only way I think we see major tweets - rather than the minor "put X town in Y seat and remove Z town from Y seat" - is if said tweaks increase COI or the stated goals of each present seat. The one example that immediately comes to mind is Denver Hispanics could want in on CD8, and would lead to major ripples across the metro. CD1 would probably grab from non-Aurora CD6, CD6 from CD4, CD4 from CD7 and 8, and CD7 from whiter suburbs of CD8.

[Oryxslayer]

https://twitter.com/DKElections/status/1407828715690770440
WTF?

tweet and pdf were deleted?

And? I don't see the problem. They picked a race closest to 50-50 and used it when aiming for competitive districts, since these districts would then come closest to matching the state's median. There's an argument of course that Colorado is not as competitive anymore as that race suggests - which is why CO07 is a Dem district and not a swingy one - but the point wasn't competition. It was to equitably match the state's partisanship.

[Oryxslayer]

https://twitter.com/DKElections/status/1407828715690770440
WTF?

tweet and pdf were deleted?

And? I don't see the problem. They picked a race closest to 50-50 and used it when aiming for competitive districts, since these districts would then come closest to matching the state's median. There's an argument of course that Colorado is not as competitive anymore as that race suggests - which is why CO07 is a Dem district and not a swingy one - but the point wasn't competition. It was to equitably match the state's partisanship.

Is a fair Kentucky map now supposed to use Beshear 2019 numbers?You know as well as I do, that this is an absurd idea.

It doesn't matter what race you pick to draw from if your goal is to equitably match the state's overall partisanship rather than competitiveness. This is a goal that was adopted by said commission early in the process. State's don't move in a bubble. If say they used 2020 partisanship, a district similar to CO07's electoral data would still appear going by their guidelines. Colorado is a Dem state, so 5-3 Dem, but Colorado isn't 100% committed to Dems so neither should the seat. Drawing a marginal seat based on a marginal result therefore fulfills these goals, since the state won't normally be as competitive as said race, and said seat therefore won't be marginal in most circumstances.

If your goal was competitiveness, then yes, you should use the most modern and all-encompassing data available.

[Oryxslayer]

Except the commission isn't even supposed to consider matching partisanship. They can use partisan data for competitive purposes and in that scenario the most obvious data set would be either Cook PVI,75/25 PVI or composite score.

The commission appears to have self-defined competitiveness as "If the statewide race is competitive, then the map should be able to swing based on said race." This definition does make some sense to adopt if your state is close-ish to the center of the nation. It was reported previously that they wanted to reflect that states overall partisanship while also fulfilling their mandate.

This is a significantly different definition from "Even If a state votes  Party +X%, there should be large numbers of marginal seats." This definition makes sense if said state is 100% safe like say Arkansas, and control is not in question. People are shy to completely subscribe to the second definition even in the best of times, since it often means artificially accepting that both parties are equivalent when one has a statewide advantage, and therefore gerrymandering the minority party opportunity seats even though the majority party is naturally dominant. See the fallout from the Arizona and Washington maps last cycle for why this is an unpopular position, even though the Arizona map retained its equitability despite the states slide towards competitiveness.

[Oryxslayer]

It appears there are no road connections in CO-3 going from Grand to Boulder so that might need to be changed. Not even sure why you would do that.

The handful of boulder precincts getting restored to the county - probably in exchange for Gilpin - seems like the most obvious adjustment that will occur in the final map. In fact, its so obvious one wonders if the west Boulder County stuff was deliberately done to provoke specific public input.

[Oryxslayer]

It seems as if the original Hispanic Buisness plan was just a ploy by some Democrats to get 5 solid seats.

https://www.coloradopolitics.com/colorado-in-dc/hispanic-business-group-calls-for-new-congressional-district-north-of-denver/article_b7f5a43c-c960-11eb-9701-430145d1eb54.html

The original map would have just had 5 purely Safe D districts. Its unlikely its a genuine hispanic play as a better Hispanic district would be South Adams and Aurora which is around 37% instead of 30% Hispanic like this. It seems as if most of said attempt worked although the commission did push for some form of competition in said district?

The map you posted is why I, speaking as a Dem here rather than an analyst, am 100% fine with the direction the commission is taking. An initially safe 5-3, or 6-2, as you note, would never be approved, and Boebert's seat has too many natural COIs to get carved into oblivion. These fantasies of the twitter-verse were in effect people more or less pissing into the wind. More likely based on COIs was a map as you present, that uses Douglas as a base and makes the fourth seat more GOP leaning. So one based out of Jefferson is a nice result. If Dems lose a 7th or 8th similar to those presented by the commission then they are likely looking at 170 seats nationwide or less, and CO is the least of their worries. Meanwhile the third remains marginal (who knows what the future holds for this seat), and whenever El Paso finally flips in 6 to 10 years then the Dems have a new target.

[Oryxslayer]

Zoom ins of the cities from the preliminary state house map. Old Maps for comparison use 2020 Pres. State House:

Denver





Northern Cities





Colorado Springs

[Oryxslayer]

State Senate:

Denver





Northern Cities





Colorado Springs

[Oryxslayer]

Couple reactions:

The addition of the Canon City region to the South Colorado State Senate seat doesn't make much sense.

I'm fairly sure the outer El Paso State House seat is illegal by road connections, and will get changes to resemble it's predecessor.

I wonder why the Ken Caryl House seat gets extended way to highlands in the southwest.

No more Rural Arapahoe + Rural Adams state house seat.

I assume there was some pop reason for swapping suburbs around between the northeastern Douglas House seat and the outer 'remainder' one.

No more parallel Weld senate seats, I think this new version with Loveland and Longmont seats works better, especially in regards to Broomfield.

The Adams and Arapahoe senate seats no longer have tails out into the county rurals east.

The Outer Jefferson senate seat seems like an obvious Dem loss, but the new Highlands Ranch senate seat seems ideal for a pickup.

[Oryxslayer]

God this map looks hideously ugly. Serious what gives ?

The commission has included in their definition, the idea of preserving cities and localities. A noble and productive intention, but the side effect is a bunch of "rough" edges. This is because cities sprawl and have tended to annex whatever they want when you go west of St. Louis, leading to some peculiar lines. It also leads to a few "remainder" rural/exurban districts, like the one in El Paso, that will probably get scrapped cause of road connection or the lack of a COI at the expense of the other seats.

[Oryxslayer]

El Paso County appears to be 10Kish over the 8 CD threshold in the new numbers, so goodbye any hope of a neat whole-county district.

[Oryxslayer]

Wouldn't be surprised if this new version gets another big change next time. There are hundreds of comments on the commission website by people upset about slicing Fort Collins out of Larimer and sticking it with the rurals.

Yeah that was a decision that just raised more questions then it answers. The simple solution would be to just cut Boulder instead and put more turf with the metro seats, but given how drastic  map a -> map b was, map b -> map c will likely be similar.