Sorted by New

Topic Contributions


Guided by the Beauty of One’s Philosophies: Why Aesthetics Matter

Thanks for sharing.  I'm hesitant about this post's thesis because I think conflating the aesthetic with the political is the reason why so many efforts at improving things failPeople end up supporting policy ideas based on empirical premises that are narratively compelling but wrong.  There are many examples, but blank slatism comes to mind.  Beyond empirical beliefs, there are broader concepts like tradition, individualism, or nationalism which IMO lack justification for being inherently good, but people have mood-affiliated themselves into worldviews and ideologies built around them anyways because of their intense aesthetic appeal.  Mixing up the two also makes for bad aesthetics.  There are exceptions but I think most propaganda is bad art.

Even aside from the epistemic effects, I'm not sure cultivating an associated aesthetic style would make effective altruism more persuasive. Some ideas have more aesthetic potential than others.  I think effective altruism does not have that much.  (I think longtermism has a lot.)  You said that effective altruism's lack of a visual aesthetic makes you less interested in it, but a bad visual aesthetic is probably worse than none, no? High modernism's core ideological commitments don't seem insane on their face but someone could reasonably look at a Le Corbusier building and think whatever worldview shares a bed with that can't be right.

In fact, it seems to me like many very successful social and intellectual movements lack a distinct visual aesthetic.  Visual symbols and slogans and recurring subject matter, yes. But I can't think of distinct visual styles for, e.g., liberalism, the Civil Rights Movement, the Enlightenment, law and economics, feminism, other stuff.

My experience with imposter syndrome — and how to (partly) overcome it

A central moral claim of effective altruism is the equal respect owed to beings regardless of what they can do for you because what matters is the capacity for joy and suffering.  That's why the drowning child hypothetical is so compelling.   The observation that some people are more talented than others is, or at least should be, only relevant to effective altruists instrumentally in service of the moral claim.

FTX/CEA - show us your numbers!

Isn't part of what's going on with FB, McKinsey, and Goldman  spending at schools that they're in an arms race  with their direct competitors for talent?   I.e. they have to spend to keep up with Alphabet/Bain/Morgan Stanley.  I don't think there's an analogue for EA - I'm sure EA groups are competing with these types of companies, but they're much more widely distributed, so the dynamics are less intense.  I don't mind spending on community-building but I would be curious to see evidence of impact (it may be available and I just don't know) since I've been on both sides of on-campus recruiting and suspect the marginal impact of money spent is ~0.  And unlike many longtermist projects, the data to assess impact should be pretty accessible?

Free money from New York gambling websites

Re: this -

  • Q: Does this entail any financial risk?
  • A: No. As long as you only bet the bonus money and don't gamble your own funds, there's no way to end up with less than you started. I recommend making one bet on each site, then withdrawing your money and closing your accounts immediately after each bet ends.

I don't think this is accurate because most of these free bet offers require that you gamble your own funds to unlock the bonus (in VA, all but one require this - BetMGM, Barstool, and Unibet), and none of these sportsbooks lets you withdraw the bonus.  There is a long discussion coming to this conclusion in KaseyShibayama's comment thread but I really think it's worth amending the actual post since I know there are EA's who have lost money on this.

EA Fundraising Through Advantage Sports Betting: A Guide ($500/Hour in Select States)

ETA: the T&C of the sportsbook promos I use as examples below have changed since I posted, so don't rely on this comment to figure out what the promos are/what the optimal strategy is

Just wanted to pipe up because I think this may not be clear to people reading this post — unless you fully hedge, this is not a risk-free opportunity, and if you are pursuing the recommended strategy in this post there’s a very good chance you will lose money.  I wanted to comment since I know at least three EA’s who have lost a couple thousand dollars taking advantage of these promos.  One very smart person who skimmed this post thought it was describing “free money,” and unless I’ve misunderstood something (very possible!) it’s not — the better way to view this opportunity is, “the E(V) of legal sports betting is always negative, but with these promotions, it’s positive.”  If you have a lot of money such that you can afford to be risk-neutral, then hell yeah and go forth, but if you don’t, it’s rational for you to be risk-averse and this might not be a good idea.

If it’s unclear how this is risky, consider that the most common type of free bet promo is one where (1) you unlock the free bet only if you lose your first real-money bet and (2) your free bet doesn’t return the stake.  In this case, as the post points out, the optimal strategy is to choose long odds on both bets.   This is probably obvious, but that means you are more likely than not to lose your money.

To take full advantage of the offer you’d be making uncorrelated bets on multiple sportsbooks, but that will lessen the risk, not eliminate it.  I live in DC — the four “free bet” options in Virginia are Barstool ($1000), BetMGM ($1000), Caesars ($1500), and Unibet ($500).  (As of this comment Fanduel no longer has the free bet promo at least in VA.)  Barstool, BetMGM, and Unibet unlock the free bet only if you lose the first (real-money) bet.  Caesars unlocks the free bet upon deposit.  Of the four, only Barstool returns the stake on the free bet.  So the optimal strategies for the four in order of first (real-money) bet and second (free) bet are:

Barstool:  Long, short

BetMGM:  Long, long

Caesars:  Short, long

Unibet:  Long, long

I’m too lazy to plug in sample numbers but you can hopefully deduce from this that (1) you have a good chance of ending up positive, since it’s optimal to go short on Barstool and Caesars, but (2) it’s still pretty easy to end up in the red.  And that’s with 2/4 offers that come as close as you can get to free money (Barstool and Caesars, though note Barstool has a -200 line limit, limiting how short you can go).

Of course if you’re risk-averse you can choose a different, non-E(V)-maximizing strategy, but I think some people might understandably just follow what’s recommended here.

Sorry if this was obvious to everyone, I just know some people at least were surprised that you could lose money on this.  I don’t think the post is misleading, though I probably would have made clearer at the top that this is not even close to a risk-free endeavor, especially considering the marketing language around these promos already can make them seem otherwise.

Matt Levine on the Archegos failure

Credit Suisse was just one of I think 5-10 different counterparties Archegos had.  As Matt Levine talks about in previous newsletters on Archegos, some banks, e.g. Goldman, JP Morgan, came out better than others, some even ended up in the black.  I wonder if that's solely attributable to the sequence in which the counterparties sold off the underlying assets - Goldman and JP Morgan initiated the sell-off - or if some of the other counterparties also managed the risk better.  If so, it would make for an interesting comparison - what factors made one huge heavily-regulated organization pull this off and another mess it up?