Comment by cryptonector | Hacker News Reader

Comment by cryptonector | Hacker News Reader

Parent Full thread

Gerrymandering is self-limited.

The party in power (in a State government) can either maximize the number of seats it has, or make seats safe. It cannot do both at once!

Maximizing seats -> minimizing the margin of victory in each district -> increasing likelihood of losing its majority in a future election!

Making seats safe -> maximizing margin of victory in each district and minimizing the number of seats by concentrating the majority's voters -> minimizing margin of victory in the popular houses (of the State legislature and Congress) -> increasing likelihood of losing its majority in a future election!

That is, no matter how you redraw district borders, as long as population per-district is roughly even (in each State), then you cannot get a long-lasting advantage.

This will probably be surprising to some here, but it's clearly true.

This means that Gerrymandering is a much smaller evil than one might think.

I would further argue that Gerrymandering is not an evil at all. It is part and parcel of the Republican system of government in the U.S.:

- the Senate, especially when Senators were selected by State governments, and still now due to its 6-year terms and staggered elections, modulates the popular House

- first past the post leads to a small number of parties and reduces party power (you vote for a person, not just a party as in the case of proportional representation systems)

- gerrymandering also modulates popular will

- juries modulate law and prosecutors

- the president is selected by the Electoral College, which also modulates popular will

- the States have police power, which modulates Federal power

- the Courts modulate Congress and the Executive branch

These things were well thought out by the Founders! They were arrived at by compromise. They weren't accidents born of ignorance of mathematics or science. This system has served us well. Changing it requires broad consensus, which requires strong arguments. So far I've yet to see any decent arguments against this structure.

Saying gerrymandering is self-limiting because the party has to choose how to allocate its power is like saying sexual assault is self-limiting because the perpetrator can only use one orifice at a time.

Gerrymandering gives an absolute advantage. At the core of it, you are destroying part of the votes of your enemies, either by bundling them in one district, or completely destroying their power by cracking them.

Flawed argument, but good on you for noticing the tradeoff. Yes, there are two types of gerrymandering, with almost exactly opposite mechanics.

Unfortunately, the rest of your comment isn't even sophomoric...the US political structure is about sloppily conceived and aimless as your constant use of the phrase "X modulates Y".

I was actually paraphrasing a majority opinion by Justice O'Connor.

I thought gerrymandering is the art of creatively drawing the lines such that all of the opposing constituents are squeezed into one or a few "throwaway districts", thus preserving comfortable majorities in the remaining districts.

How is this self-limiting? Also, I don't quite see how it maps to the rather abstract phenomenon you are talking about?

Edit -- wikipedia seems to match up with what I was remembering: https://en.wikipedia.org/wiki/Gerrymandering

Cryptonector's comment makes very little sense. It's an obvious advantage if you can move the boundaries between two seats in order to go from 80-20/50-50 to 75-25/55-45.

It's somewhat frustrating though when you can see that one area is 200k people and another is 50k people but the party negatively affected (someone will be) by re-drawing the boundaries to be "fairer" will scream bloody murder and claim Gerrymandering.

It's only a temporary advantage.

If you maximize seat safety, you minimize the number of seats you have because you bunch up all your voters. This means that the opposition need only flip a few districts to win back a majority (and then redistrict to their liking!).

If you maximize the number of seats you have then you spread your voters thin. This makes it easier to win in a wave: with many districts close to 50/50, you only have to flip a few percent of the voters state-wide or nation-wide and you win. And that's exactly what's been happening. We've had a number of wave elections recently.

Folks, I was paraphrasing a majority opinion by Justice O'Connor (retired). Shaw v. Reno[0], IIRC.

Downvote all you want, but all you're doing is showing your bias and ignorance. This isn't the sort of subject where you can reliably determine the truth value of any statement a commenter might make. You should be more open minded. I get the impression that if the tables were flipped you'd flip too, that the disgust you express at redistricting (or the Electoral College, or...) is merely convenient and insincere.

EDIT: And the tables have flipped over the years, haven't they. Democrats ran the table and redistricted accordingly for over 60 years between 1932 and 1994. I'm OK with that. I'm OK with Republicans doing the same now, and I will be OK with it when Democrats next get to do it again. Redistricting doesn't seem to have the effect you all seem to fear. In spite of redistricting, 2006 and 2008 happened, did they not? The 60 years that Republicans spent in the wilderness weren't due to redistricting, they were due to the long-lasting cultural effects of the Great Depression. It's reasonable to suspect that without the Great Depression we would have seen a much higher frequency of majority party switches in the 20th century. I think a switch frequency of 1 every 20 years is a reasonable guess. I can live with that.

[0] https://en.wikipedia.org/wiki/Shaw_v._Reno

Interesting take, but it assumes a high margin of error in the voting preferences of certain constituents, which may not be true, which likely makes it not self-limiting enough.

It assumes the opposition can sway some percentage of the population in a wave in one case (maximizing seats) or that they can put up a few really good candidates that can win on personal attributes and policy, not policy alone, in the other (maximizing seat safety).

We've certainly had wave elections, have we not? 1920, 1932, 1994, 2006, 2010... They're getting more frequent, perhaps because redistricting may not be maximizing seat safety as much? (I'm not sure, I've not and will not spend my cycles on analyzing redistricting history.)

Reply on news.ycombinator.com