POLITICS DEC. 25, 2021 Democrats Are Doing Weirdly Well in Redistricting

[oxmixmudd]

In other words, "Democrats do it too"

Source: https://nymag.com/intelligencer/2021/12/democrats-are-doing-weirdly-well-in-redistricting.html

The Democratic House majority was supposed to die in redistricting. For months now, pundits and political forecasters have predicted that Republicans could win back the House next year without flipping a single voter. After all, the GOP controls far more state governments than the Democrats, and this is a post-Census year, when states redraw their congressional maps. Republicans boast sole authority over the boundaries of 193 congressional districts, while Democrats command just 94. Given the slimness of Nancy Pelosi’s majority, several analyses projected that GOP cartographers would generate enough new, safe “red” seats to retake the House through gerrymandering alone.

This has been a foundational premise of much of my own commentary. And it’s an assumption that’s animated the progressive movement’s push for a package of democracy reforms that would, among other things, forbid partisan redistricting.

But it’s starting to look wrong.

The new House map is more than half finished. And in many states where maps haven’t been finalized, the broad outlines are already visible. Taken together, the emerging picture is far more favorable for Democrats than most anticipated. As of this writing, it looks like the new House map will be much less biased in the GOP’s favor than the old one. And according to at least one analyst, there is actually an outside chance that the final map will be tilted, ever so slightly, in the Democrats’ favor.

...

There are a few reasons why things didn’t work out as progressive pessimists had feared. One is that — contrary to partisan stereotypes — Democratic trifectas have arguably mustered more ruthless party discipline in redistricting than Republicans have. Illinois, Oregon, and New York have all pursued aggressive partisan gerrymanders that have subordinated the job security of some incumbents to maximizing the overall number of Democratic-leaning seats. By contrast, Texas Republicans took the opposite approach, opting to fortify their incumbents’ hold on power, at the cost of leaving 13 Democratic-leaning seats on the map. Meanwhile, many red states have no room to improve on existing gerrymanders.

To be sure, blue states have probably left more gerrymander-able seats on the table than red ones, simply because some of the nation’s most Democratic states have outsourced redistricting authority to independent commissions. Fortunately for Team Blue, California’s nonpartisan commission is poised to finalize a quite pro-Democratic map. As of this writing, California’s House map is likely to feature 44 seats to the left of the country, and eight to its right. If Democrats boasted full control over California redistricting, they probably could have produced a 50-to-2 Democratic gerrymander. But still, not a bad haul.

...

According to Wertheimer’s calculations, if both of those gerrymanders are rolled back, then it is actually possible that the “tipping point” seat in the final, nationwide map will be one that was slightly more Democratic than the nation as a whole in 2020. Which is to say: The House map could end up having a tiny pro-Democratic bias.

This is by no means the likely outcome. But its plausibility underscores a basic fact: The biggest threat to the Democrats’ House majority in 2022 is no longer Republican gerrymandering but rather, the combination of the opposition party’s inherent turnout advantage in midterms and Joe Biden’s dismal poll numbers.

© Copyright Original Source

It's wrong regardless of who does it, and it should be illegal. I hope one day soon it is simply a law that the districts can't be gerrymandered to keep one party in power even if a significant majority of the public votes against that party. How to define gerrymandered? You set error bounds based on the number of districts to the population - with a minumum number of districts (this keeps the granularity from being too large 10 districts == 10%, 20 == 5% and so on), If ANY* popular majority difference larger than the granularity of the districting plan can't get their choice elected, it is gerrymandering and must be redone.

*no matter how that majority is distributed across the districts, a majority > the granularity will shift to the majority vote party. I tend to think a minimum number of districts is 25** (4% granularity), so that as long as at least 52% of the population votes for a candidate, they WILL be the candidate elected.

** there may need to be exceptions to that district count for extremely low population areas

[rogue06]

It's wrong regardless of who does it, and it should be illegal. I hope one day soon it is simply a law that the districts can't be gerrymandered to keep one party in power even if a significant majority of the public votes against that party. How to define gerrymandered? You set error bounds based on the number of districts to the population - with a minumum number of districts (this keeps the granularity from being too large 10 districts == 10%, 20 == 5% and so on), If ANY* popular majority difference larger than the granularity of the districting plan can't get their choice elected, it is gerrymandering and must be redone.

*no matter how that majority is distributed across the districts, a majority > the granularity will shift to the majority vote party. I tend to think a minimum number of districts is 25** (4% granularity), so that as long as at least 52% of the population votes for a candidate, they WILL be the candidate elected.

** there may need to be exceptions to that district count for extremely low population areas

If 52% of the voters in a state vote for someone, they win no matter how the Congressional districts are laid out.

If 52% of the voters in a district vote for someone, they win no matter how the district was laid out.

Are you conflating gerrymandering with the Electoral College and presidential elections?

[CivilDiscourse]

It's wrong regardless of who does it, and it should be illegal. I hope one day soon it is simply a law that the districts can't be gerrymandered to keep one party in power even if a significant majority of the public votes against that party. How to define gerrymandered? You set error bounds based on the number of districts to the population - with a minumum number of districts (this keeps the granularity from being too large 10 districts == 10%, 20 == 5% and so on), If ANY* popular majority difference larger than the granularity of the districting plan can't get their choice elected, it is gerrymandering and must be redone.

*no matter how that majority is distributed across the districts, a majority > the granularity will shift to the majority vote party. I tend to think a minimum number of districts is 25** (4% granularity), so that as long as at least 52% of the population votes for a candidate, they WILL be the candidate elected.

** there may need to be exceptions to that district count for extremely low population areas

To be honest, I think you need to re-word this. I've read it like 3-4 times, and I'm uncertain what you are trying to say.

My best interpretation is that you are trying to create a districting system where the districts vote as a block, and the blocks then tally up to the election of a single candidate, which isn't how it works.

[oxmixmudd]

To be honest, I think you need to re-word this. I've read it like 3-4 times, and I'm uncertain what you are trying to say.

My best interpretation is that you are trying to create a districting system where the districts vote as a block, and the blocks then tally up to the election of a single candidate, which isn't how it works.

Thanks for giving me the benefit of the doubt. So, gerrymandering can help to keep the balance of power in both the local state legislature and congress for a party that does not have the popular vote when taken as a sum across the entire state. It can do that by packing a district with a particular kind of voter (e.g. African American), diluting their part of the vote in other districts and keeping that party's representation by the states elected officials out of sync with the aggregate vote of the entire population of the state, or by the opposite, making sure a particular voting block is broken up to the point they rarely if ever can get enough votes to get the candidate(s) of their choice elected. My wording is confusing because there is on the one hand the cumulative effect of the total number of candidates across all the districts and the singular effect of the abstract 'candidate', singular' that represents what is happening wrt each individual district.

What I am proposing is that one look at what happens in the aggregate in terms of the total power of a given party in the local legislature and in congress and the capacity of a shift in the popularity of that party on the potential or expected shift in the representation of the aggregate set of elected officials for the state, either locally, or nationally in congress. If it is possible, to the granularity of the district number, for a party to remain in power as a group of elected officials in their local legislatures or their representation in congress even if across the state (in the sum of all the district votes) the majority of that state's population is voting for the other party, again to the granularity of the district number, then the districts are gerrymandered.

[CivilDiscourse]

Thanks for giving me the benefit of the doubt. So, gerrymandering can help to keep the balance of power in both the local state legislature and congress for a party that does not have the popular vote when taken as a sum across the entire state. It can do that by packing a district with a particular kind of voter (e.g. African American), diluting their part of the vote in other districts and keeping that party's representation by the states elected officials out of sync with the aggregate vote of the entire population of the state, or by the opposite, making sure a particular voting block is broken up to the point they rarely if ever can get enough votes to get the candidate(s) of their choice elected. My wording is confusing because there is on the one hand the cumulative effect of the total number of candidates across all the districts and the singular effect of the abstract 'candidate', singular' that represents what is happening wrt each individual district.

What I am proposing is that one look at what happens in the aggregate in terms of the total power of a given party in the local legislature and in congress and the capacity of a shift in the popularity of that party on the potential or expected shift in the representation of the aggregate set of elected officials for the state, either locally, or nationally in congress. If it is possible, to the granularity of the district number, for a party to remain in power as a group of elected officials in their local legislatures or their representation in congress even if across the state (in the sum of all the district votes) the majority of that state's population is voting for the other party, again to the granularity of the district number, then the districts are gerrymandered.

If I'm understanding you correctly, the ultimate goal would resemble a legislature that has a party* makeup similar to the population proportion. (I'm using party because we operate in a party system.) I don't think i'm summing it up quite accurately, but I think that is the overall gist of the proposal. Also, Once you start chunking a whole up for individual mini-elections, the aggregate vote tally becomes, not only just a bit of trivia, but also a potentially misleading statistic. It makes the assumption that turnout between districts will be equal. This isn't necessarily true,

I've never been sold on that idea of legislature balance = voter balance. It's not a "natural" outcome of redistricting. Though, districting is never "natural" to begin with. It's creating artificial boundaries, and there's almost always going to be someone's opinion driving those. If each district represented the same distribution of as the population as a whole (say statewide race for each legislator, or randomized population distribution), then the majority party would hold nearly every seat in the legislature. To change from that outcome, you have to start deciding where the best places to draw the lines are. And you have to decide what that outcome should ideally look like.

[oxmixmudd]

If I'm understanding you correctly, the ultimate goal would resemble a legislature that has a party* makeup similar to the population proportion. (I'm using party because we operate in a party system.) I don't think i'm summing it up quite accurately, but I think that is the overall gist of the proposal. Also, Once you start chunking a whole up for individual mini-elections, the aggregate vote tally becomes, not only just a bit of trivia, but also a potentially misleading statistic. It makes the assumption that turnout between districts will be equal. This isn't necessarily true,

No - what I'm looking at is not outcome based. I could do some simulations I suppose to try to make it clearer. But perhaps I could come up with an extreme example that could illustrate the point. Gerrymandering is done because it works - it helps shift that balance of power away from the voters and into that hands of the party currently in control of the local legislature or the states representation in congress. So what I'm looking at is how, in the aggregate, voting shifts in populations that normally vote for the ruling power can affect that party's representation in aggregate. If the portion of the population that normally provides the in power party its success shifts to start voting for the other party, the districts, and the ultimate representation of the people, should also reflect that same shift. Gerrymandering allows the ruling power to ride out or even defeat the effect of such shifts in popularity.

I've never been sold on that idea of legislature balance = voter balance. It's not a "natural" outcome of redistricting. Though, districting is never "natural" to begin with. It's creating artificial boundaries, and there's almost always going to be someone's opinion driving those. If each district represented the same distribution of as the population as a whole (say statewide race for each legislator, or randomized population distribution), then the majority party would hold nearly every seat in the legislature. To change from that outcome, you have to start deciding where the best places to draw the lines are. And you have to decide what that outcome should ideally look like.

What I'm proposing can still result in 100% representation by the ruling party, or anything in between. What it stops is allowing that same ruling party to stay in power (in the aggregate) even when the population of the state has left them behind for the other party.

[CivilDiscourse]

No - what I'm looking at is not outcome based. I could do some simulations I suppose to try to make it clearer. But perhaps I could come up with an extreme example that could illustrate the point. Gerrymandering is done because it works - it helps shift that balance of power away from the voters and into that hands of the party currently in control of the local legislature or the states representation in congress. So what I'm looking at is how, in the aggregate, voting shifts in populations that normally vote for the ruling power can affect that party's representation in aggregate. If the portion of the population that normally provides the in power party its success shifts to start voting for the other party, the districts, and the ultimate representation of the people, should also reflect that same shift. Gerrymandering allows the ruling power to ride out or even defeat the effect of such shifts in popularity.



What I'm proposing can still result in 100% representation by the ruling party, or anything in between. What it stops is allowing that same ruling party to stay in power (in the aggregate) even when the population of the state has left them behind for the other party.

Ok, it seems I just am not understanding what you are saying then.