Roy Moore vs Doug Jones, Alabama Senate Race

[Timsup2nothin]

This must only be true to a degree... I don't think they'd be in favour of the Alabama congressional seat situation as depicted here. A 6:1 seat advantage isn't a bipartisan gerrymander, and you wouldn't expect one in such a polarised, majority conservative state.

Depends which "they" you ask. I'd bet that Terri Sewell, who has represented the 7th district for the past three terms, thinks it's fine. Keep in mind that in Alabama it would probably be possible to gerrymander EVERY district to the Republicans.

 

[Arwon]

I reckon that person would probably like some company - looks like it'd be easy to do at least two, and possibly three, safeish Democratic Party seats from that voter distribution.

 

[Arwon]

Yeah, looks like Terri Sewell isn't happy with the gerrymander: https://twitter.com/RepTerriSewell/status/915239657616629760

And in fact seems to have been the person to introduc the act to restore the Voting Rights Act, partly in response to the current wave of racial gerrymanders - https://sewell.house.gov/media-cent...g-rights-advancement-act-restorethevote-speak

 

[Timsup2nothin]

I reckon that person would probably like some company - looks like it'd be easy to do at least two, and possibly three, safeish Democratic Party seats from that voter distribution.

Keep in mind that the distribution you are looking at is from the recent special election...which is a genuine aberration. Last year the closest any of those six republicans came to losing was in the second district...where Roby "only" won by ten points. The rest won by thirty points or more, if there was even an opponent willing to take the beating.

 

[Timsup2nothin]

Yeah, looks like Terri Sewell isn't happy with the gerrymander: https://twitter.com/RepTerriSewell/status/915239657616629760

LOL...you are thinking that she isn't happy because she says she isn't?

C'mon, seriously?

 

[Arwon]

Even so, looking at the county-level results in 2016, they could plausibly have a district based mostly around Birmingham and another based on Montgomery, Tuscaloosa and that belt of Democratic voting heavily black counties in between them, and both would be pretty Democratic. There's enough concentrated areas of Democrat vote that a 5:2 split would be easily achieved with relatively contiguous districts.

 

[Timsup2nothin]

Even so, looking at the county-level results in 2016, they could plausibly have a district based mostly around all of Birmingham and another based on Montgomery, Tuscaloosa and that belt of Democratic voting heavily black counties in between them, and both would be pretty Democratic. There's enough concentrated areas of Democrat vote that a 5:2 split would be easily achieved with relatively contiguous districts.

Perhaps. The 5-2 split with no competitive districts is not a goal anyone supports though...other than the seven elected representatives. Truthfully, the map as it stands is about the best that can be done in a state where one party has a two to one advantage. With turnout like we had Tuesday at least a few districts would have been competitive. Even though the Republicans would have held them all at least they'd have had to work at it.

 

[Arwon]

I'm just skeptical that Sewell or the Democrats got a say in the Alabama districting, yeah (at state level they've sued to overturn districting of the Alabama state congress)

 

[Arwon]

Also with the extreme geographical (ie racial) polarisation of Alabama, a set of mostly safe districts is probably what you'd expect from impartially drawn, political boundary and community-of-interest based electorate boundaries. One of the many nefarious effects of segregation combined with single-member first-past-the-post electorates.

 

[Timsup2nothin]

I'm just skeptical that Sewell or the Democrats got a say in the Alabama districting, yeah (at state level they've sued to overturn districting of the Alabama state congress)

Alabama is pretty far gone, so not really an ideal example, but yeah.

 

[Arwon]

Oh I fully take your point about bilateral and often racially-focused gerrymanders in general, I just think Alabama was clearly unilateral with racial representation as mere figleaf to stave off any court challenges.

 

[Broken_Erika]

It's Alabama. What do you expect?
I mean, Alabama....I could post the South Park Alabama Man Action Figure video again, but i'm not sure if the mods would appreciate having the same video posted every other page. Even if it does some up Alabama's Ideal Candidate..

 

[JollyRoger]

Moore should argue that he won the electoral vote in Alabama by 6 to 1 and should be seated retroactive to when America was last great.

 

[Broken_Erika]

More like he'll argue that he is God's preferred candidate and the will of God take precedence over the will of voters because Jesus told him so.

 

[JollyRoger]

God certainly has shared Moore's taste for underage girls.

 

[Broken_Erika]

God Be Praised!

 

[dutchfire]

Math time:

2016 2-party presidential vote in Alabama: N=2047802
2016 GOP presidential vote in Alabama: K=1318255
1/7th of total presidential vote (average district): n=292543
14/th of the total presidential vote (needed to win average district): a=146271

If we would randomly draw n voters from the total voting population N, the number of GOP voters x is given by a hypergeometric distribution with mean nK/N = 188322 and variance
var = nK/N * (N-K)/N * (N-n)/(N-1)=57506

Simplifying to a normal distribution with that mean and variance (law of large numbers), the probability of a Republican to lose a 'random' district would be something like 0.23232 (if my online calculator isn't failing me), which would correspond to 1.6 expected R losses.

This math is a bit funky (and voters are not distributed randomly geographically), but having 1 Democratic seat looks consistent with fair districting.

 

[hobbsyoyo]

It's perfectly correct as it stands; you should leave it. She does more damage (by inciting right-wing response) than she causes directly.

I'm with you in your main point; she should recognize that the best way she could serve the party is to stay out of sight--recognize and then do.

I also agree that Bernie is already too old. In any case, the next candidate should not be a progressive. It should be someone who can unite centrists and progressives in a wave that can keep Trump in check, sweep Trumpism away. That's the most urgent political need at present. Then Dems can go back to bickering.

So you're saying we need another Obama. Agreed.

 

[hobbsyoyo]

This must only be true to a degree... I don't think they'd be in favour of the Alabama congressional seat situation as depicted here. A 6:1 seat advantage isn't a bipartisan gerrymander, and you wouldn't expect one in such a polarised, majority conservative state. Bipartisan gerrymanders are probably also easier in big states with lots of districts to play with like California.

It is true to a degree. You can only point to a handful of states (and really mostly cities in particular) and say this is going on. You can however say that partisan gerrymandering that favors Republicans went on in practically every state they controlled in 2010-2012.

It's probably fair to say redistricting was always intended to be political process but I do not think that process was meant to be gamed as thoroughly as it has.

Math time:

2016 2-party presidential vote in Alabama: N=2047802
2016 GOP presidential vote in Alabama: K=1318255
1/7th of total presidential vote (average district): n=292543
14/th of the total presidential vote (needed to win average district): a=146271

If we would randomly draw n voters from the total voting population N, the number of GOP voters x is given by a hypergeometric distribution with mean nK/N = 188322 and variance
var = nK/N * (N-K)/N * (N-n)/(N-1)=57506

Simplifying to a normal distribution with that mean and variance (law of large numbers), the probability of a Republican to lose a 'random' district would be something like 0.23232 (if my online calculator isn't failing me), which would correspond to 1.6 expected R losses.

This math is a bit funky (and voters are not distributed randomly geographically), but having 1 Democratic seat looks consistent with fair districting.

Which isn't the same thing as saying the districts are fair. But thank you for educating me!

 

[JollyRoger]

Math time:

2016 2-party presidential vote in Alabama: N=2047802
2016 GOP presidential vote in Alabama: K=1318255
1/7th of total presidential vote (average district): n=292543
14/th of the total presidential vote (needed to win average district): a=146271

If we would randomly draw n voters from the total voting population N, the number of GOP voters x is given by a hypergeometric distribution with mean nK/N = 188322 and variance
var = nK/N * (N-K)/N * (N-n)/(N-1)=57506

Simplifying to a normal distribution with that mean and variance (law of large numbers), the probability of a Republican to lose a 'random' district would be something like 0.23232 (if my online calculator isn't failing me), which would correspond to 1.6 expected R losses.

This math is a bit funky (and voters are not distributed randomly geographically), but having 1 Democratic seat looks consistent with fair districting.

fair enough. But this show a 6 to 1 split based on the 2018 special election numbers where the party that got the most statewide votes would still only have won one district. This shows how the gerrymandering can help keep power even when the overall populace is against yu. Run your math again using the 2018 numbers to see what the expectations of losing a district are.