Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[silverspawn]

Oh and the ending is pretty cool

[silverspawn]

Oh no Andrew Yang is peddling the "Biden can't win" narrative. I'm disappointed

At least now if he does get blown out, it has a silver lining that I need to be less disappointed in everyone who says it.

But it just seems so silly. I just checked the last 19 polls, and it comes out at less than a 2% lead for Trump in the average, at a point before they're even the nominees, while the fundamentals are looking amazing. "Biden has no chance" is not a rational thing to conclude from this data.

538 has had really good success with predictive models for decades, and they've included fundamentals in all of them, and they've looked at whether models without them do better, and they don't.

In 2020, polls were showing an 8 percent lead for Biden, and I heard less of "Trump has no chance" than I'm hearing now.

[silverspawn]

With Nate fired from ABC news, we probably won't get a forecast this year. I'm following Nate's substack and I don't think he's in the mood to do it himself. He is justifiably annoyed by all this stuff and seems much more interested in talking about poker and sports

[silverspawn]

Andrew of course has a personal interest in saying Biden has no chance since he now has a third party, but I legit thought that he'd be a sufficiently rational guy as to not let that influence him

[silverspawn]

GPT-4 > Andrew Yang. SAD! BABY!

[faust]

Andrew of course has a personal interest in saying Biden has no chance since he now has a third party, but I legit thought that he'd be a sufficiently rational guy as to not let that influence him

What does rationality have to do with it? If it is to his personal advantage to say it, then the "rational" thing is to do so regardless of whether he believes it.

[silverspawn]

Andrew of course has a personal interest in saying Biden has no chance since he now has a third party, but I legit thought that he'd be a sufficiently rational guy as to not let that influence him

What does rationality have to do with it? If it is to his personal advantage to say it, then the "rational" thing is to do so regardless of whether he believes it.

If he's deliberately lying, then yeah. But I don't think that's likely.

[faust]

Andrew of course has a personal interest in saying Biden has no chance since he now has a third party, but I legit thought that he'd be a sufficiently rational guy as to not let that influence him

What does rationality have to do with it? If it is to his personal advantage to say it, then the "rational" thing is to do so regardless of whether he believes it.

If he's deliberately lying, then yeah. But I don't think that's likely.

I'm just a bit confused on the consequentalist take on this. In terms of outcomes, the best thing (from Andrew Yang's perspective at least, and from yours if you prefer him over Biden) would be to lie, right? (Assuming that you don't believe that Biden can't win)

So then, if you think he's a candidate that follows your ethics, you should want him to say that Biden can't win in either case, right? Either he believes and says it, or he doesn't believe it and still follows consequentalist logic and says it.

Then how can you be disappointed if he says it? What am I missing here?

[silverspawn]

Yeah, I see how that question makes sense.

The thing is that I think deliberately lying for better outcomes is something that ethical consequentialists almost never do, although there are certainly some exceptions. (Like SBF, maybe.)

Doing stuff like that is sometimes called "naive" consequentialism or utilitarianism. The idea is that you trust too much in your personal calculation or model and ignore the fact that lying has all these difficult to compute but severely negative downstream effects, and that empirically, it usually doesn't work out. I remember Eliezer Yudkowsky talking about this particular point (why don't you do all sorts of unethical stuff in service of saving the world), and his response was basically that this just almost never works out. He mentioned Knut Haukelid as the only historical case of a ruthless altruist where it seemed like it was worth it in hindsight.

I think in practice, most ethical consequentialists don't even consider lying as an option. I certainly don't. Like the theoretical justification is that out-of-model variables tend to systematically make lying worse than you think, but practically, you just don't think about it. In this case, the idea that Yang could be lying on purpose didn't even occur to me.

You could construct a scenario where I would lie, but you'd have to try pretty hard to come up with extreme variables.

[silverspawn]

if Sam Bankmann Fried (SBF) did indeed take EA axioms seriously (which I think is reasonably plausible), that's a great example. He did various shady stuff with FTX, which in a naive calculation all works out. EA got a shit ton of money, and who cares if some random people somewhere get screwed over; the benefit probably outweighs the harm by a lot.

... except that it all blew over and did immense damage to EA. So in retrospect, lying was absolutely not a good idea. And the planning fallacy is so hard to overcome and everyone will underestimate the chances that things go horribly wrong somehow, it just tends to be better to adopt a strict rule against lying or similarly unethical behavior

[faust]

This is true, but I guess I figured in this particular case (lying about one's personal beliefs), the dangers are not that hard to estimate. It's going to be pretty hard for anyone to catch you in that lie, and even if they do, you could just say you changed your mind.

When you go into politics, I don't think you can be successful by always being completely honest, so these sorts of considerations should come naturally. I was recently at a local party meeting and even then I had to tackle the question in my mind of whether I should be honest about a topic that I knew wouldn't go over well with that particular crowd.

[silverspawn]

When you go into politics, I don't think you can be successful by always being completely honest, so these sorts of considerations should come naturally.

Yeah, that might be true. I wasn't considering the politician variable because he said it in his personal podcast rather than during a rally or TV interview, but maybe I should have.

[Awaclus]

Yeah, I see how that question makes sense.

The thing is that I think deliberately lying for better outcomes is something that ethical consequentialists almost never do, although there are certainly some exceptions. (Like SBF, maybe.)

Doing stuff like that is sometimes called "naive" consequentialism or utilitarianism. The idea is that you trust too much in your personal calculation or model and ignore the fact that lying has all these difficult to compute but severely negative downstream effects, and that empirically, it usually doesn't work out. I remember Eliezer Yudkowsky talking about this particular point (why don't you do all sorts of unethical stuff in service of saving the world), and his response was basically that this just almost never works out. He mentioned Knut Haukelid as the only historical case of a ruthless altruist where it seemed like it was worth it in hindsight.

I think in practice, most ethical consequentialists don't even consider lying as an option. I certainly don't. Like the theoretical justification is that out-of-model variables tend to systematically make lying worse than you think, but practically, you just don't think about it. In this case, the idea that Yang could be lying on purpose didn't even occur to me.

It should occur to you 100% of the time that a politician could be lying on purpose. Lying works out great a lot of the time, there's just an obvious selection bias that conceals all the times when it does from public awareness.

[silverspawn]

Yeah, I see how that question makes sense.

The thing is that I think deliberately lying for better outcomes is something that ethical consequentialists almost never do, although there are certainly some exceptions. (Like SBF, maybe.)

Doing stuff like that is sometimes called "naive" consequentialism or utilitarianism. The idea is that you trust too much in your personal calculation or model and ignore the fact that lying has all these difficult to compute but severely negative downstream effects, and that empirically, it usually doesn't work out. I remember Eliezer Yudkowsky talking about this particular point (why don't you do all sorts of unethical stuff in service of saving the world), and his response was basically that this just almost never works out. He mentioned Knut Haukelid as the only historical case of a ruthless altruist where it seemed like it was worth it in hindsight.

I think in practice, most ethical consequentialists don't even consider lying as an option. I certainly don't. Like the theoretical justification is that out-of-model variables tend to systematically make lying worse than you think, but practically, you just don't think about it. In this case, the idea that Yang could be lying on purpose didn't even occur to me.

It should occur to you 100% of the time that a politician could be lying on purpose. Lying works out great a lot of the time, there's just an obvious selection bias that conceals all the times when it does from public awareness.

Makes sense.

... aren't you a politician?

[Awaclus]

... aren't you a politician?

Yes! I feel pretty uncomfortable with the idea that I could be caught lying without plausible deniability (probably more so than is rational; experienced politicians get caught lying all the time and their reputation seems to be mostly fine typically), so I almost exclusively say things that are either literally true or unfalsifiable. I also intrinsically value honest communication with smart people, and I don't have the kind of platform where I would have a large number of people who aren't that smart paying attention to what I say. But if I am communicating with a potential voter, then of course I optimize for getting them to vote for me, not for having them believe true things.

[Awaclus]

And it's not super uncommon for me to come across new information that introduces some nuance to some of my political talking points and be like "oh that's interesting, I'll keep this in mind so I can make better decisions, and whenever I expect idiots to see my posts, I'm going to pretend that this information doesn't exist".

[silverspawn]

La Heine (2005)

Is this movie a masterpiece? Man I don't know. YMS gave it a 9 and I don't have any technical complaints. But I'm also not interested in watching a bunch of morons being morons for 90 minutes. Everyone in this movie is just so stupid.

(It's about french teenagers who have issues with the police.)

Also there was some symbolism involving a cow and the eiffel tower. Did it mean something? I don't know. I wasn't paying that much attention; I was getting bored so I played smash on the side.

3/10

[silverspawn]

Wait it's 1995 not 2005 my bad

[silverspawn]

well maybe it's not rational but it's rationalizing!

[silverspawn]

I've long thought that I probably play some aspects of poker horribly wrong while spending most time optimizing stuff that I already do quite well. This is how it usually goes; you focus on details but miss the big stuff. Which is very inefficient. The problem is, I didn't know what those aspects are.

Now I think I've found out. The thing I do horribly wrong is play as the big blind. I have no idea what to do, so I usually just fold to avoid the situation, and apparently that's a massive mistake. And it's probably one of the most important positions to play well.

[silverspawn]

I shudder to think how much meticulously earned equity I've thrown away by having no clue when to defend as the big blind. Here's hoping I can improve a lot now.

[silverspawn]

No Grammarly, I know that cutting words like "simply" makes you sound more confident, but I don't think you can change "Simply put, [...]" to "Put, [...]"

[silverspawn]

Put, it's because it doesn't sound right!

[silverspawn]

yes a whole new world of big blind play is opening up to me

[silverspawn]

So way back, I remember having an argument with Awaclus on whether or not it makes sense to help other players get better at games like Dominion.

I think this is a case where I think my original inclination was correct, i.e., it doesn't make any sense. But to elaborate on this, I think what's going in is mostly:

- People are social creatures. Helping others feels nice and appropriate.
- People are STATUS HUNGRY. Helping others CONFIRMS YOU AS HIGH STATUS since you KNOW STUFF THAT THEY DON'T KNOW and if your teaching is appreciated you receive SWEET AFFIRMATION that's incredibly YUMMY and you need MOAR.
- Arguing that you shouldn't help people is incredibly autistic.
- It's incredibly difficult to help anyone get better in Dominion anyway since it's incredibly talent based relative to just about every other game I know, so this entire conversation is entirely theoretical anyway.
- (Yes this undermines the stated purpose of the forum. I think I've said this before. But the forum is good in plenty of other ways! I mean its main function has always just been that it brings like-minded people together. The goal may not be achievable, but they journey is fun, and as every buddhist knows, the journey is what matters.)
- Incredible is an incredibly frequent word in this incredible list.

But putting all of this aside, the stated goal of helping people get better at a game like this doesn't make any sense I don't think. There is no public good if people play better. All you're doing is... well nothing since you can't help people at dominion, but all you would be doing if it worked would be to give people in your audience an advantage over people not in your audience.

The dynamic sharpens a bit in poker. The potential to improve is much greater, and the zero-sum-ness is even more obvious since you're playing for money against other people, so others improving is actively bad for you. Even if everyone including you improves equally it's actively bad since it takes away potential for you to improve relative to others.

I mean the obvious economic truth is just that coaching is a market. If the discipline is sufficiently popular, it's a market for money; if it's not, it's a market for status validation, but the principles are basically the same. I remember having this conversation with someone in Prismata and he argued that everyone improving is good for him since then he'll play against stronger players. The degree to which I don't buy this reason in hindsight exceeds my lack of understanding of the last two dominion expansions. But it's a common lie that's mostly socially upheld. You probably want to believe that you're doing a good thing being a coach.

Chess also a huge coaching scene. It seems to be an outlier in that coaching actually has a lot more money in it than professional play, which is the opposite of how you'd think it works.