foxes.bsky.social 🦊🏳️‍🌈

You are correct. I don't know why I brought strategy into this when not needing to vote strategically is why I'm such a fan of approval voting in the first place.

a good upgrade pick if you like how they sound, but they cost 2x more.

IMO they're tuned way better than the Sony products, but I think their tuning is pretty polarizing?

Probably not worth it, though.

The 1000XM4s are *insane* value right now if you only care about ANC. You can get them for $160 right now. I don't love how they sound, but if you buy from somewhere w/a good return policy, you can always return them if you feel similarly.

The JBL Tour One M3s are a

Omg my bad.

You don't have to worry about codecs if you're planning to hold on to your phone for much longer.

If you're only in the market for over-the-ear headphones and want to spend a fortune, the Sennheiser HDB 630s are probably the nicest on the market right now for sound tuning and quality.

higher-than-optimal threshold, the electoral result that maximizes overall voter satisfaction just won't be represented, because on the margin, the penalty for not polarizing your platform really does matter.

Under FPTP with primaries, you have intra-party caucuses and electoral fusion, so someone aligned with the Working Families Party beliefs might win a Democratic Party primary and be listed in the general under a joint WFP/Dem label. Same outcome in the end.

But in both cases, until you hit a

between TCP and first vote preferences in Australia (which seems to keep going up?) suggests you guys still have some issues w/rt polarization are overshadowing specific voter preferences.

The same is true here. Trump's platform was seen as relatively moderate by voters in the NYT/Siena polling in the lead up to the 2024 election.

It's just the case that the incentives still reward being those who are further from the median voter than what maximizes voter satisfaction.

The gap

It's "approval voting but more complicated" more than it's "RCV but more complicated"

Game out the incentives for strategic voting under STAR. There are ~no benefits. Under approval, you have some incentive to bullet vote. STAR fixes that but it's otherwise pretty similar.

What do you consider good? My preferred voting system prevents extreme and unpopular candidates (e.g. Trump) from winning against candidates who are closely aligned with voter beliefs. My nightmare is one that incentivizes adopting positions and passing policy that voters oppose. RCV is the latter.

Lander (likely the approval and Condorcet winner) losing and having to endorse Mamdani is a problem with the game theory of RCV.

4 vs 5 is supposed to matter. If you like Mamdani and Lander equally, you can assign them the same score.

It happens elsewhere, too. Ireland's rates are exceptionally low (humans count the ballots and the ballots are filled w/numbers rather than scanned bubbles, which seems to help) but even Ireland has spoilage problems.

In US primaries, it's pretty bad: electionconfidence.org/wp-content/u...

No.

Eliminating center squeeze and making strategic voting near useless are huge advantages for a voting system.

But RCV tends not to live up to the ideal assumptions.

RCV is so polarizing that you shouldn't expect ranking more to change much about elections: races still coalesce around a tiny number of people on very different sides.

It was always Lander, Mandani, and Cuomo in the NYC race, for example.

It ends up not making a difference.

Having a primary system is equivalent to having ranked choice voting where you can only rank two candidates. Having RCV in primaries lets you rank more, but fundamentally, ranking two gets you 90% of the way there in practice even under ideal assumptions for RCV.

you buy something last-gen second hand. And if you do care about speed and you trust and want a OnePlus, you get a much faster and more efficient phone.

Somehow, for the first year in my life, I actually don't have strong opinions on this.

I would say, if you can wait a little bit, phones with the new Snapdragon chips are launching, notably the OnePlus 15R and 15.

Even if you don't care a ton about speed, it should help with prices on Swappa if

Primaries and STAR both let voters express preferences.

compulsory voting. Australia has compulsory voting at least, but needing it is unfortunate.

I'd call RCV w/Coombs an improvement over FPTP+primaries, but in any world where esoteric systems are in the cards, but not Australian IRV.

but it's hard to prove a negative, so I'll just reiterate that I don't feel that the theoretical properties change much: you only gain clone independence when moving to RCV, but IIA makes that less beneficial than you'd think.

- Like FPTP w/primaries, it sucks w/rt participatory/GOTV incentives w/o

4/n:
- Spoilers are still a huge problem / no IIA.

- Strategic voting is still extremely effective in any race where RCV has the chance to improve outcomes.

- Extremely similar VSEs even in a world where ballot spoilage isn't a problem. (Wolk 23)

- I am not sure what properties you care about,

3/n

- Australian ballots are spoilt at substantially greater rates than the US

- Australian politics are *highly* polarized (evidence: I follow you and see your tweets 😛). If you type "center squeeze" you'll see a million polsci papers explaining why: it's for the same reason that FPTP polarizes.

2/n

The promises of RCV simply do not materialize. See e.g. www.nowpublishers.com/article/Deta...

Since you mention Australia:
- Donkey voting is measurable and genuinely problematic in Australia, and the obvious solution (randomized ballot order) causes other problems (Horiuchi & Lange)

1/n
Since your hands are Australian: My tweet is referring to American FPTP, where we do primaries w/sore loser laws. But the statement is (to a lesser degree) true.

RCV is unambiguously worse than American-style FPTP. We regularly try RCV in different states and cities and it always gets repealed.

Boring answer: elastane can just do that and you won't be able to do anything about it. 20 C is a pretty safe temperature.

Long answer requires what is essentially an interview about your underwear & laundry to understand to explain a fairly small amount of variance. Not the best use of your time.

(Or I can demand a partner do the same thing, you get the point)

It breaks vote secrecy. I can buy a vote by telling someone to cast a very specific ballot.

spoiled ballots are a huge issue w/minority and low income voters under RCV.

Even uglier than all of that: you have really bad participation criterion effects. See e.g. Peltola vs. Begich: Begich lost because he got 5,200 more votes! (tbf, I am glad he lost. but still!)

don't actually appear.

At the same time, RCV makes election data much harder to work with (it's harder to understand voter intentions, nearly impossible to audit elections in a privacy-preserving way, etc., which makes it a political non-starter: it tends to just get repealed), etc.

Beyond that,

It's fundamentally equivalent to doing a hunch of rounds of first past the post voting (hence why it's synonymous w/ instant runoff voting) and so the majority of the theoretical issues with FPTP don't actually go away. Empirically, some get worse, and as predicted by theory, the proposed advantages