Is America A Natural GOP Gerrymander?

[Slander and/or Libel]

Your concerns are certainly valid (not that you need me to tell you that). I've toyed with the following idea (it's in very early stages, not terribly well thought out yet):

States with 5 or fewer representatives would each consist of one multi-member district from which all representatives would be drawn.

States with 6 or more representatives would draw multi-member districts of no fewer than 3, no more than 5 representatives per district, subject to reasonable constraints on the drawing of the districts, preserving COIs, etc.

That might not be enough to really dilute the effect in the most highly concentrated urban areas, but it would suffice for most.

[angus]

Angus that's precisely the problem. I have to look up a district every time it's referenced and it does take time. I just don't have to do that when discussing Canadian or British districts.

Really, you don't have to look up stuff when you hear "Clackmannanshire:  16560 yes votes, 19036 no votes" and the like?  I think most of us do.

As an experiment I just typed in Clackmannanshire in the Google search field.  The first hit was a Wikipedia article.  The second was the Clackmannanshire Council web page.  The third was some sort of archival stuff.  Etc., etc.  Then I clicked on the news tab, leaving the search field alone.  The first hit was "Scottish independence: Clackmannanshire first to declare."  Linking to that article gave me some numbers, but no other information.

Now, let's type TX-3 in the search bar.  Not much better.  Here the first hit is a pair of shoes made by Puma.  The second is a cooling fan for a central processing unit, the Cooler Master Hyper TX3.  If I instead type "house district TX-3" I get a Wikipedia article as the first hit, but the second is a direct link to Sam Johnson's web page. 

Yeah, okay, I'm convinced that TX-3 is no better than Clackmannanshire, but it seems no worse.  I guess it's whatever you're used to, and I'm used to the XX-nn notation, so it seems normal.   

[The Mikado]

As an aside, I wish you Americans would name your districts. Do you know how bloody annoying it is to hear TX-3 or whatever and have to look it up on a map every time? Just call it Dallas-East or something

Hearing Sam Johnson's district described as Dallas-East is just bizarre to me, given that the district is almost totally located in the suburbia north of Dallas.

[memphis]

Suppose there was a fair congressional map in the 2012 election. Would the GOP still have won a majority of seats?

I'd recommend this DKE diary.

The author puts me back in TN-9, where I belong. If only it were true!

[DC Al Fine]

Angus that's precisely the problem. I have to look up a district every time it's referenced and it does take time. I just don't have to do that when discussing Canadian or British districts.

Really, you don't have to look up stuff when you hear "Clackmannanshire:  16560 yes votes, 19036 no votes" and the like?  I think most of us do.

As an experiment I just typed in Clackmannanshire in the Google search field.  The first hit was a Wikipedia article.  The second was the Clackmannanshire Council web page.  The third was some sort of archival stuff.  Etc., etc.  Then I clicked on the news tab, leaving the search field alone.  The first hit was "Scottish independence: Clackmannanshire first to declare."  Linking to that article gave me some numbers, but no other information.

Now, let's type TX-3 in the search bar.  Not much better.  Here the first hit is a pair of shoes made by Puma.  The second is a cooling fan for a central processing unit, the Cooler Master Hyper TX3.  If I instead type "house district TX-3" I get a Wikipedia article as the first hit, but the second is a direct link to Sam Johnson's web page. 

Yeah, okay, I'm convinced that TX-3 is no better than Clackmannanshire, but it seems no worse.  I guess it's whatever you're used to, and I'm used to the XX-nn notation, so it seems normal.   

Sure, There's no reason why you shouldn't have any trouble so long as you have a reasonable command of geography.

As an aside, I wish you Americans would name your districts. Do you know how bloody annoying it is to hear TX-3 or whatever and have to look it up on a map every time? Just call it Dallas-East or something

Hearing Sam Johnson's district described as Dallas-East is just bizarre to me, given that the district is almost totally located in the suburbia north of Dallas.

I was pulling numbers/names out of my butt.

[muon2]

As an aside, I wish you Americans would name your districts. Do you know how bloody annoying it is to hear TX-3 or whatever and have to look it up on a map every time? Just call it Dallas-East or something

Hearing Sam Johnson's district described as Dallas-East is just bizarre to me, given that the district is almost totally located in the suburbia north of Dallas.

I was pulling numbers/names out of my butt.

As long as gerrymandering is legal and used extensively in the US, named jurisdictions won't really be very useful. TX-3 is at least a compact CD, unlike many others that would be near impossible to describe (help me with my home state of IL ). But even with TX-3 one has to choose how to name it: Plano (about a third of the CD) or Plano/McKinney or TX-Collin (most of it is in that one county, but not exclusively) or Metroplex NE.

[angus]

When I type "PA-6" in the google search engine, I become very educated about DuPont's Zytel High-Performance Nylon, a polycaprolactam otherwise known as Zytel PA6^®.

"PA-16" teaches me about the Piper PA-16 Clipper, an extended fuselage propeller airplane, modeled more or less after the PA-15 Vagabond.  (Of course, a quick search using "PA-15" will teach you all about that model.)

In honor of this turn of events, I propose that we rename my congressional district, formerly known as PA-16, to the more user-friendly "Lancsterberks" after the three counties it passes through (Lancaster, Chester, and Berks).  Note that it does not contain all of these three counties, but only part of them.  However, it is the only congressional district in pennsylvania that contains at least part of these three counties and no others, so the name will be unabiguous.  (For example PA-7, or should I say, the district formerly known as PA-7, contains parts of these three counties, but also contains parts of Delaware and Montgomery counties, so it might be known as "Lancsterberks Delmont."  Ah, this is easy.

Alternatively, we could rename them from the more offensive XX-nn system to the less offensive Representative Name system.  Mine would be called Pitts, after congressman Joe Pitts.  Given that incumbents are so safe in these districts, they would not have to be renamed often.  Certainly Pitts could have kept its name for 20 years, although given that the old boy is 75 years old, we might have to come up with a new name soon.

As another alternative, we could use the shape system.  PA-16 is shaped like this:



I'd call that district "Screaming Man with Bird Perched on his Lip and Firecracker on his Forehead"

Of course, that is only easy with regularly-shaped, not terribly Gerried districts like this one.  What could we come up with for the name of the neighboring district, formerly known as PA-7?

[Sol]

Of course, that is only easy with regularly-shaped, not terribly Gerried districts like this one.  What could we come up with for the name of the neighboring district, formerly known as PA-7?

Philly West and Japan?

[angus]

Not bad.  Now that you mention it, I see a guy (philly west) putting a foot against a person in front of her, a Geisha, presumably to tighten her corset.   Philly west has the hands behind the back and is about to grab the strings of the corset, and is screaming at the Geisha, "Suck in stomacho.  Tomodachi San not accept you in this condition.  Hai!"  Philly west also has some weird pony tails flying up in the air, like Pluto ears.  (I mean Pluto the dog, not the planet.)

[Brittain33]

Baby Bullwinkle

[angus]

PA-16 should be "Reading & Lower Susquehanna Valley East".  

I think you have made the point very well with that post.  Let's just stick with PA16.  

Man, what a gerried set of districts.  I'm assuming that Reading was attached to PA16 because it absorbs the 'hood but not so much that it makes Pitts vulnerable.  The Horse'n'Buggy country in Eastern LanCo is detached from the rest because it is reliably Republican and can attached to Bullwinkle (suburban Philly) in order to make PA7 safe GOP.  (Old Order Amish are apolitical and don't vote, but the goyyim who live among them have been reliably Republican since at least 1860.)  As for PA6, it incorporates quite a bit of Horse'n'Buggy country as well as the White Flight areas outside Reading proper so it is naturally Republican.  It might not be if Reading were included, so that's probably why Reading was put in PA16 instead.

One big clusterfuck guaranteeing at least three safe R seats for congress.  That's what I call progress.

[The Mikado]

Getting back to the original thread question, a "fair" map is an impossibility, but I think we're seeing that a hypothetical fair map would probably still return a GOP congress in a D+2 or D+3 year.  The current gerrymander, though, leads me to suspect that Congress wouldn't flip unless the election were something like D+6 (not entirely random guess, but guess not supported by numbers).

[angus]

Getting back to the original thread question, a "fair" map is an impossibility...

Isn't it?

http://www.vox.com/2014/5/8/5695350/this-is-what-america-would-look-like-without-gerrymandering

[muon2]

Getting back to the original thread question, a "fair" map is an impossibility...

Isn't it?

http://www.vox.com/2014/5/8/5695350/this-is-what-america-would-look-like-without-gerrymandering

I also think a fair map is possible, though the splitline algorithm in the link is not the way to get there. I would go a step further than the video in the link which suggests the commission route, and instead crowdsource the map. Set up criteria and tools and let the interested public decide.

[traininthedistance]

Yeah, splitline isn't "fair" at all due to how it disregards natural borders and metro areas.

[angus]

I noticed that the spline ignored populated areas.  I think there are advantages and disadvantages in doing that, but I liked the commission route as well.  muon's suggestion about crowdsourcing is intriguing, but it has the disadvantage of potentially making itself for sale to the highest bidder, whereas a commission of highly-vetted, thoughtful people might be less susceptible to marketing.

[muon2]

I noticed that the spline ignored populated areas.  I think there are advantages and disadvantages in doing that, but I liked the commission route as well.  muon's suggestion about crowdsourcing is intriguing, but it has the disadvantage of potentially making itself for sale to the highest bidder, whereas a commission of highly-vetted, thoughtful people might be less susceptible to marketing.

If there's a clearly defined set of measurable criteria, then there's little to worry about marketing. Some deep-pocketed interested party could invest in a plan, only to be bested by some netizens in their parents' basements or a couple of science professors playing with numbers in their spare time.

[angus]

Ha.  Ye of goode faith.  I hope that it would work out so well, owing to netizens and bored scientists.

Of course the state lege would have to go for it.  What would be the odds of getting your colleagues in the state assembly to go along with such a plan?

[angus]

This is my district and I had absolutely no clue how gerrymandered this was. What an utter embarrassment. I would've figured at least everything south of the Main Line would be part of this district (namely Paoli/Malvern) and probably expected up northward to the PA Turnpike. Here I was complaining for months about how my state assembly district was gerrymandered because Freshman year I was one of three dorms put in a Solid Democratic district before finally switching into the Solid Republican district this year. Didn't even think the US reps from PA were this bad. Should've known since there are never any tight races in this state. This and Fitzpatrick's district up near Bucks County are two of the most competitive in the state!! Should've realized the gerrymander wasn't completely natural, but I figured it was reasonable to have slightly Republican districts just outside of 99% D Philly (especially in Delaware County moreso than Montgomery and Bucks). Boy, this is disheartening.

I'm glad you're feeling it.  You're tomorrow's state senator, or at least tomorrow's voter.  Make it change.

[traininthedistance]

Of course, that is only easy with regularly-shaped, not terribly Gerried districts like this one.  What could we come up with for the name of the neighboring district, formerly known as PA-7?

This is my district and I had absolutely no clue how gerrymandered this was. What an utter embarrassment. I would've figured at least everything south of the Main Line would be part of this district (namely Paoli/Malvern) and probably expected up northward to the PA Turnpike. Here I was complaining for months about how my state assembly district was gerrymandered because Freshman year I was one of three dorms put in a Solid Democratic district before finally switching into the Solid Republican district this year. Didn't even think the US reps from PA were this bad. Should've known since there are never any tight races in this state. This and Fitzpatrick's district up near Bucks County are two of the most competitive in the state!! Should've realized the gerrymander wasn't completely natural, but I figured it was reasonable to have slightly Republican districts just outside of 99% D Philly (especially in Delaware County moreso than Montgomery and Bucks). Boy, this is disheartening.

Obama broke 60% in Delco not only in 2008, but also in 2012; and in fact no Republican has won it since 1988.  To be fair they still have some ancestral down-ballot strength, but for federal offices it is anything but natural to make "slightly Republican districts" in the Philly suburbs- the exact opposite (slightly Dem districts) would be correct instead.

Worth noting that Bucks district is actually pretty much the only district in the state that isn't a complete abomination; for whatever reason Bucks' county lines are kept sacrosanct even when the Republicans pull out all the stops everywhere else.  And Bucks is a good deal more Republican than Delco, as well: even though it also hasn't voted R for Prez since the elder Bush, the margins have been much narrower, and the right Pub could still capture it, which just isn't true for Delco anymore.  

Also worth noting that the Philly burbs have been gerrymandered for quite some time, but the current batch of lines is arguably the worst in the country (though North Carolina and Maryland give it stiff competition).  I had the misfortune of casting my 2006 vote in this monstrosity:

[The Mikado]

Getting back to the original thread question, a "fair" map is an impossibility...

Isn't it?

http://www.vox.com/2014/5/8/5695350/this-is-what-america-would-look-like-without-gerrymandering

Well, the objectives usually listed for a non-gerrymandered USA aren't all in alignment.  "Compact districts that preserve communities of interest" doesn't necessarily jive too well with "Maximize number of competitive districts."

[traininthedistance]

Getting back to the original thread question, a "fair" map is an impossibility...

Isn't it?

http://www.vox.com/2014/5/8/5695350/this-is-what-america-would-look-like-without-gerrymandering

Well, the objectives usually listed for a non-gerrymandered USA aren't all in alignment.  "Compact districts that preserve communities of interest" doesn't necessarily jive too well with "Maximize number of competitive districts."

I don't actually care too much about maximizing competitive districts.  Some areas are rock-solid one way or another and that should just be respected.  I do care about, if not completely counteracting the effects of natural packing, mitigating their worst excesses, so that for each state a 50/50 result (measured via uniform swing, presumably) comes out to something resembling a 50/50 seat distribution.  And, preferably, that the slope as you go away from 50/50 in either direction isn't too asymmetrical.

Balance that with "compact districts that preserve communities of interest" and we're good to go.  Those two ideals are not one hundred percent compatible, but I think they can be reconciled much better than a mandate to maximize competitive districts.

[○∙◄☻¥tπ[╪AV┼cVê└]

On a related, simpler to analyze topic, the States seem to be a Republican gerrymander. The Senate naturally favors Republicans. The median state is halfway between Ohio and Florida, or R+1.5. If states voted with their PVI, and the tied state of VA split its Senators, Republicans would have a 53-47 majority.

60 Democrats means going to R+6 Georgia. 60 Republicans only relies on winning 3/4 of the D+1 states.

[muon2]

Ha.  Ye of goode faith.  I hope that it would work out so well, owing to netizens and bored scientists.

Of course the state lege would have to go for it.  What would be the odds of getting your colleagues in the state assembly to go along with such a plan?

The idea has gotten some nice media, but it will take the pressure of a citizen petition amendment to move the needle. The trick is getting the citizen groups to take off the blinders of a sole alternative of a CA-style commission drawing the map.

The core language might look like this:

The Commission shall provide to the public data and tools to create Legislative and Representative districts. The Commission shall accept maps for the redistricting of Legislative and Representative districts.

The Commission shall evaluate all submitted maps according to criteria as implemented by law and assign each map a score. The Commission shall eliminate maps by a majority vote of the members appointed that fail to meet federal and State law and shall eliminate maps that are substantially the same as other maps of equal or better score. The Commission shall give to the Senate and House the maps for Legislative Districts with the best three scores.

If the Senate or House has failed to file a redistricting resolution with the Secretary of State by the deadline, the Secretary of State shall certify the redistricting map that received the best score from the Commission.

[○∙◄☻¥tπ[╪AV┼cVê└]

Ha.  Ye of goode faith.  I hope that it would work out so well, owing to netizens and bored scientists.

Of course the state lege would have to go for it.  What would be the odds of getting your colleagues in the state assembly to go along with such a plan?

The idea has gotten some nice media, but it will take the pressure of a citizen petition amendment to move the needle. The trick is getting the citizen groups to take off the blinders of a sole alternative of a CA-style commission drawing the map.

The core language might look like this:

The Commission shall provide to the public data and tools to create Legislative and Representative districts. The Commission shall accept maps for the redistricting of Legislative and Representative districts.

The Commission shall evaluate all submitted maps according to criteria as implemented by law and assign each map a score. The Commission shall eliminate maps by a majority vote of the members appointed that fail to meet federal and State law and shall eliminate maps that are substantially the same as other maps of equal or better score. The Commission shall give to the Senate and House the maps for Legislative Districts with the best three scores.

If the Senate or House has failed to file a redistricting resolution with the Secretary of State by the deadline, the Secretary of State shall certify the redistricting map that received the best score from the Commission.

It's only fair that some Republican states disarm themselves before any more Democratic states do.