[testdelacc1]

Why is there always a both sides-er in these discussions?

FWIW, one party generally deferred to nonpartisan commissions to draw boundaries to avoid gerrymandering. So one “side” did far more than propose a solution, they did the right thing even when the other side wasn’t.

Gerrymandering is the worst example to pick when you’re pushing both-sides-bad.

[paulsutter]

Great! So please explain what algorithms are used by these "nonpartisan" commissions?

[testdelacc1]

I’ll reply in good faith even though I detect sarcasm in your comment.

Generally nonpartisan commissions prioritise contiguity and compactness. There is an element of “I know it when I see it” because you’re trying to avoid both packing (packing minority voters from disparate areas into one) and cracking (distributing a minority district like Salt Like City into its 4 neighbouring districts, ensuring the city can’t vote for … whoever cities generally vote for).

So there is a human element involved, but these commissions generally do a reasonable job. You know how we know? States that move from nonpartisan to partisan commissions cause a dramatic change in the results of the next election. If the nonpartisan was biased like you imply with your air quotes, we wouldn’t observe that effect.

Also there are algorithms to draw fair districts without needing human judgement. See this paper[1] that expounds on one such algorithm.

1 - Swamy, R., King, D. M., & Jacobson, S. H. (2022). Multiobjective optimization for politically fair districting: A scalable multilevel approach. Operations Research, 71(2), 536–562. https://doi.org/10.1287/opre.2022.2311

[paulsutter]

Thank you, I really appreciate it. Found a nonpaywalled link:

https://optimization-online.org/wp-content/uploads/2019/03/7...

It is exciting to see research, but this algorithm has never been used by any commission anywhere (unless it's been adopted since publication! I would love to hear)

Can you find any algortithms actually used by these "nonpartisan" commissions you mention? Or even could you explain how nonpartisan participants are selected? (even if their commissions dont use any objective methodology)

I would truly love to see a way out of this mess, but I've been hoping for more than “I know it when I see it”

[unselect5917]

They're not.