>If Ethereum or Bitcoin had a 40%-50% chance of doubling or better by end of year, every investor in the world would be putting all their money in it.
Not if they thought it had an equal chance of losing more than half it's value. If you see crypto as still somewhat in its early stage of price discovery where the supply is very well known but the demand could potentially increase a lot if more mainstream investors get involved (especially US retail investors once some bitcoin-based ETFs become available) then there's a good chance the price could bubble up past 100k. There's too many unknowns to say EMH should apply just because it's gone a bit mainstream.
>Not if they thought it had an equal chance of losing more than half it's value
Shouldn't this be losing all its value? If I can buy something for 100 and there's a 50% chance it will be worth 200 or 50 in a year, that's an ev of 125, and a good deal (ok, not for *all* my money, but there should be enough money around that would take an opportunity like that)
Jack's claim was "would be putting all their money in it". Kelly betting says you bet to maximise E(log(wealth)); when you're betting all your money, therefore, 50% of double/50% of halved is exactly even (50% of double/50% of -100% has E(log(wealth) = -infinity; you should never make a bet with a nonzero probability of total ruin, since then you can't make any more bets).
Kelly betting approximates maximising E(wealth) if you're not betting very much of your net worth (linear approximation), but you need to consider the nonlinearities when "all their money" is invoked.
Inflation is still low though according to the data such s CPI and barely budged despite all this printing. The expansion of the money supply is mostly going to assets and paying off debts or saved, not being spent on consumer goods as much or in such a way that it would create inflation in terms of rising price levels. Even if consumer spending does go up, this does not mean inflation if companies can increase production to meet demand and not raise prices.
I suppose the people who are out there doing research on men's apparel might choose the brands to sample in such a way that the "hedonic adjustment" calculation moves in one direction for a few months and then the other direction for a few months, but they're not going to be able to do anything systematic for an extended period, and they'd have to somehow collude across many dozens of categories to even produce a systematic short-term effect.
The full quote from the remote work article is "Employees can apply for up to 12 months but only those deemed to be in “exceptional circumstances,” will have their request fulfilled.", which says the opposite of what you're saying it says.
Firstly, how do you know that? Secondly, even that's a very different standard from Scott's prediction criteria of "widely allows remote work, no questions asked".
As a counter point to the above (commenting here for the Google wfh point)
1) his 80% is the probability of Biden having a higher than 50% approval rating. That seems quite plausible to me.
2) I work at Google and my reading of that question is that anyone who wants to could work from home, full time, even outside of their office city. I think Google management has been pretty explicit / public about not committing to that - Id rate the probability <5%. Which doesn't mean that some people on the margin won't be able to work remote / not come back for an extended period by negotiating with managers, just that this won't be the norm.
He's not saying Biden has, or will have, an approval rating of 80%. He's saying that there's an 80% chance of Biden having an approval rating of 50% or better on Jan. 1 2022.
The wokies don't like Biden, that's true. Non-woke Democrats do like him (that's how he won the primary), and swing voters like him (that's how he won the general). He's popular with regular people who don't yell at each on Twitter.
How is that? The election results don't show anything about polling - just about the difference between the polled population and the voting population. "Approval rating" isn't about voting though - it is just *defined* as the results of these polls, so there's literally no way for polling error to exist.
Anyway, the election results were off by less than 5 points everywhere, and the current approval rating shows a 14 point margin, so I think that anyone who has properly taken the lessons of the past few elections should accept that Biden's positive favorability rating is "real", whatever that means.
If it's defined as the results of those polls, it's sort of a problem, right? Phone polls recently seem to have a response rate of something near 1%. Approval rating that's the result of polling the self-selected 1% of the population seems like a meaningless number.
This is also why the election results were so wildly off. If you just take that 1% and try to weigh for all demographic factors, you can't assign enough weight to the group of people who just won't answer the phone poll, because they are not in your data at all! (I know there was at least one pollster who tried, though, by asking questions like: "do you think your neighbors will vote for Trump?")
That seems far too low. As a 78 year old man, Biden has a 4.7% chance of dying just in the next year according to this table https://www.ssa.gov/oact/STATS/table4c6.html ; we'd need to adjust that for his apparently-good health, but I doubt he's got a 95% chance of living to November 2024.
Hmm... stacking the probabilities over the next four years, it's a little over 20%... and adjusting it a little downward, so about 15%? I suppose I was underestimating it, because I've never heard "Biden is likely to die" from anyone who isn't obviously in my outgroup, like Amy here.
In addition to his apparent health, you have to take into account the fact that he has regular access to the best medical care that money can buy. Presumably, he will be able to intervene very early and aggressively against any health issue that might arise.
Well, many people were happy to forecast that Trump would keel over dead some time in office, as he was so unhealthy, old, and ate a poor diet, but he's still alive. I am willing to assume that Biden may manage to live another four years to the age of 82 in November 2024. There has been some speculation about his cognitive health but there is nothing concrete about his physical health.
Possibly more likely is that in a year or two he may resign due to those cognitive difficulties to hand over to Kamala, but prognosticating his death does seem a little premature.
He didn't say Biden would die, he said it seemed like more than a 5% chance, which seems obviously true going by those tables, unless you think he's three times healthier than the average septuagenerian.
Note that I'm not predicting Biden's approval rating will be 80%, I'm saying there's an 80% chance Biden's approval rating will be above 50%. Currently it's 54%.
"Independent but the Left is worse" and "The Bay Area is a hellish sea of NIMBY hypocrisy" are two viewpoints that I rarely get exposed to on these polyonymous codices! You may consider my mind duly blown
The fact that you think 538 conducts polls indicates that you are uninformed enough on this topic that I can safely disregard your opinion. You should probably bother to read some of their reporting on the topic before so widely and confidently expressing your opinions on the matter.
From the intro: "Some predictions about my personal life, or that refer to the personal lives of other people, have been redacted to protect their privacy."
"Vitamin D is generally recognized (eg NICE, UpToDate) as effective COVID treatment: 70%"
Isn't this counter to what you previously said on Vitamin D/COVID? Have you changed your mind, or do you expect the consensus to come to an incorrect conclusion? Or just a typo?
I'm also curious about this, and if you've changed your mind about it, I think it would be worth adding a note to that effect to the original post (or if you don't have time I can at least add a comment there).
>25. Greater than 66% of US population vaccinated against COVID: 50%
50% seems too low to me, what's your reasoning here? Already more than 50% of adults have received at least one dose, and it looks like vaccines will be approved for children too soon.
There will be a sorta trail off at some point as you start to hit the subset of people who don't want to get vaccinated for various reasons.
Notably, if you look at Israel, which has led the curve of vaccination, it seems to have trailed off around 60%: https://ourworldindata.org/covid-vaccinations . The UK also seems to be leveling off.
If covid starts to reduce its infection rate enough, and enough regulations are relaxed, I won't be surprised if that reduces the number of people who want to get vaccinated further, and there's a definite chance of news events, akin to the J&J blood clots, but even more sensationalized, which impact the general public's attitude more.
I don't think Scott's prediction is too bad given how much Israel and the UK appear to be leveling off in their vaccination rate. I'd personally give it more than 50%, but I think 50% is defensible.
Israel's numbers are complicated by the fact that children are not yet eligible to receive the vaccine. If you look at what percentage of the population that _is_ eligible has received it, I think you'll find something closer to 90%.
Which is why I'd like to see Scott clarify that prediction: is that of the entire US population, or of the subset of the population that is eligible to receive the vaccine, as 12/31/2021?
The UK is levelling off because the people who got their first dose in Q1 now need a second dose, so nearly all the supply is being allocated to second doses, with very little left over for new first doses. There are still plenty of people under 45 who aren't eligible for a first dose yet but really want one.
Yeah, switching the chart to "Vaccine doses" rather than "People vaccinated" doesn't show any leveling off, which suggests its down to devoting capacity to second doses. Anecdotally, I got my vaccine this week and the center was running pretty much at full capacity, so I don't think demand is the limiting factor yet.
Israel does show a leveling off for total doses too, which does seem to be due to lack of people applying, but as others mentioned a lot of this is due to children not being vaccinated yet: around a third of Israelis are under 16, so a 60% vaccination rate is actually most of those eligible.
I was thinking the same thing at first, but then realized that the adults/children thing causes extra complications. The NYTimes is saying that 80% of Americans are 16 or over. While I think they're likely to approve vaccines for ages 12-16 in a few months, it seems less clear that they'll approve them for younger people, so there's a reasonable chance that only 85% of Americans are eligible before December. So the question is basically whether more than ~80% of eligible Americans will eventually get vaccinated.
Since we're having a major slowdown in the adult vaccination at 50%, it's definitely not obvious that we'll get to 80% by the end of the year.
Maybe a lot of it really is the inconvenience and hassle for some people and general unawareness for others, rather than specific unwillingness, and maybe they'll gradually develop "get-out-the-vote" tools that work for these groups.
Also, maybe Novavax will get approved, and the people who are worried about adenovirus/mRNA "reprogramming their genes" will get that shot. (I actually heard someone specifically say he was waiting for Novavax for that reason!)
According to a recent poll, only 69% of Americans say they want to get vaccinated. That number could go up as people learn the vaccine is safe. But it could also go down, as COVID gets better and people stop feeling it's as urgent.
Individuals likely change their minds back and forth a lot in ways that you can't see as % changes on those surveys (because they mostly cancel each other out). There only has to be a brief window where a person approves of the vaccine in order to go out and get it and then they're vaccinated forever even if they change their mind again. But also most consumer research is bunk because people don't actually do what they say they're going to do. Maybe people answer those surveys according to perceived tribal stances but when their own skin is on the line they act differently. But the fact that 50% already got it makes me want to assign >50% probability to 66%.
Oroxylum indicum's active ingredient, oroxylin A [1], appears to be a relatively understudied stimulant of the dopamine reuptake inhibiting variety [2].
I had to look it up, and it has several names including such soft and emollient ones as MIDNIGHT HORROR and BROKEN BONES 😀
It's used in traditional medicine, and this paper doing a round-up on it indicates that it will do everything short of ironing your shirts and cooking your dinner:
"Various parts of the plant are used in Ayurveda and folk medicine for the treatment of different ailments such as cancer, diarrhea, fever, ulcer and jaundice. Recent in vivo and in vitro studies have indicated its anti-inflammatory, antiulcer, hepatoprotective, anticancer, antioxidant, photocytotoxic, antiproliferative, antiarthritic, antimicrobial, antimutagenic and immunostimulant properties.
Roots are sweet, astringent, bitter, acrid, refrigerant, anti-inflammatory, anodyne, aphrodisiac, expectorant, appetizer, carminative, digestive, anthelmintic, constipating, diaphoretic, diuretic, antiarthritic, antidiabetic and febrifuges. Tonic is useful in dropsy, cough, sprains neuralgia, hiccough, asthma, bronchitis, anorexia, dyspepsia, flatulence, colic, diarrhea, dysentery, strangury, gout, vomiting, leucoderma, wounds, rheumatoid arthritis and fever. Root bark is used in stomatitis, nasopharyngeal cancer and tuberculosis. Leaves are used as stomachic, carminative and flatulent. Leaf decoction is given in treating rheumatic pain, enlarged spleen, ulcer, cough, and bronchitis. Mature Fruits are acrid, sweet, anthelmintic, and stomachic. They are useful in pharyngodynia, cardiac disorders, gastropathy, bronchitis, haemorrhoids, cough, piles, jaundice, dyspepsia, smallpox, leucoderma and cholera. Seeds are used as purgative. Dried seed powder is used by women to induce conception. Seeds yield non-drying oil used in perfume industry. The seeds are ground with fire soot and the paste is applied to the neck for quick relief of tonsil pain. The seeds are used in traditional Indian Ayurvedic medicine, included in famous tonic formulations such as Chyawanprash. Bark decoction is taken for curing gastric ulcer and a paste made of the bark powder is applied for mouth cancer, scabies and other skin diseases. The medicated oil of O. indicum in sesame oil base instilled into ears mitigates the pain in otitis."
See, this is why every time I try to learn about medicinal plants I run into a roadblock. Every plant I research supposedly does 12 different things! And every single one is a diuretic!
Really looking forward to the Oroxylum self-reports and the culture wars essay! I predict that your scores for these predictions will surpass all previous years.
You absotively-posilutely need to specify some kind of method or measure for 5. in advance, because it's going to depend a lot on (i) woolly estimation of numbers (ii) judgments about what constitutes "BLM" (iii) needing a dataset for this with a track record of year-to-year regular methodology
I think it could be an effective prediction as it is on the basis of damage *claims* of $250 million due to 'BLM protests,' rather than a specific pinning-down of what would constitute such damage.
Most news articles about last year's protests said "caused $X in damages". I'll judge this correct if a trustworthy news source says there were protests that caused $250 million in damage, unless some other news source says they're wrong, in which case I'll use my best judgment.
you need to define what you mean by "vaccine effectiveness." for instance, my daughter still got mild chicken pox, localized to her calf and a one-day fever, even though she was vaccinated. does this mean the vaccine wasn't effective?
Yes, exactly. If your implicit definition in #30 is "variant makes vaccines <=50% effective at preventing severe illness and/or death in fully vaccinated people" then 40% probability of that happening this year seems way too high-- if that is your definition I would really be interested to understand how you arrived at that probability.
>I've read and reviewed at least two more dictator books
I'm sure you already have plenty of options, but you might find Professor Archie Brown's book The Myth of the Strong Leader interesting in this regard:
I haven't yet read it myself, but I greatly enjoyed and would highly recommend Prof. Brown's books The Rise and Fall of Communism (now on sale on Kindle for $2!) and The Human Factor: Gorbachev, Reagan, Thatcher, and the End of the Cold War.
Yeah, I plan to put Yang at or near the bottom of my ranking just based on his platform still containing Basic Income garbage. Not sure how many other New Yorkers see it my way but I'd wager it's not going to help him.
I don't know what to feel about Yang's chances. I'd love to see him get the job, if only to see how his bright optimistic technocratic policies ram head-on into the wall of New York bureaucracy and institutional inertia, not to say entrenched and rampant corruption and crony politics. Sorry, Andrew!
On the other hand, there are other candidates I feel would do better - Stringer and Adams, for two. I was rather surprised that the news stories were all about the Democratic candidates, but it seems like the Republicans couldn't even catch a cold in NY governmental politics and non-party/independents are just novelty candidates with no hope at all. So it's going to be a Democrat getting elected. Yang is the frontrunner, but I dunno - I always have a feeling that people who were part of the bureaucracy/local government previously have an advantage, hence why Stringer and Adams (and maybe Garcia and Donovan, although they're both low in the polls, and I don't know if being the former counsel to de Blasio is going to help or hurt Wiley) seem to me to have a better chance when it goes to the polls with the ordinary voters, especially if they're using ranked choice voting this time.
So Yang could win the Democratic nomination but with five candidates on the ballot, the public might decide "hey, that guy OtherCandidate is the one who sorted out my problem with ApplicationForm" and go for them as Number One choice instead.
I said this last year too, but you’re way too bullish on Starship. Starship might reach space this year, but reaching ORBIT requires a fully functional Super Heavy booster, which is on version 1 “production pathfinder” right now (Starship itself is on version 15, and has still exploded every time it has flown). And Musk has some wild ass plan for catching the thing in midair with a hook on the launch pad.
My best guess for the end of the year is that Starship is flying some successful high altitude tests with non-explosive landings, but Super Heavy is at the “short hop” phase at best.
I also think 60% is optimistic, but do keep in mind that most of the Starship failures so far have been during part of the landing sequence, it is very possible that a test flight reaches orbit before either stage can be successfully recovered, in fact IIRC Musk has said that this will likely be the case. Still I'd personally put this prediction at more like 40%, rising to Scott's 60% if we extend the time-frame from Jan 1st to something more like April or May 1st 2022 (so a year from now rather than by the end of this year).
I don't think you've refuted Gbdub's point. Let's assume that Starship itself is done, finished, working perfectly, and completely ignore any landing requirements. To get to orbit, Starship still needs a working Super Heavy booster. The first Super Heavy production prototype was only assembled a few weeks ago: it'll take more prototypes, and plenty of development and integration work, before the full stack is working.
Scott's figure of 60% probability of Starship reaching orbit is, I think, about right for a window from now to the end of 2025. For the probability of reaching orbit by the end of 2021, I would put the probability at 2%. If you (or Scott) are amenable, I would be happy to bet on this, taking odds somewhere between these extremes for Starship in orbit by 31 Dec 2021. 10-to-1, say: you put up $X, I put up $10X, and the winner takes the pot. Interested?
I'm not the original commenter, but I'd be willing take that bet if you're still up for it. Say my $10 to your $100? Or higher, if 10 bucks isn't enough incentive.
I'll also offer a secondary bet of 1-to-1 odds that it makes it to orbit by end of 2023 (I think this is fair to you, if you give end of 2025 a 60% chance, but let me know).
Let me know if want to accept either, and we can work out exact terms.
Sure! Your $10 to my $100 sounds good to me: with those amounts, I'm happy to trust you for it, rather than worrying about an escrow account. Though I appreciate that you may feel differently about $100. We can discuss this.
I'll pass on the secondary bet: that one's close enough to my estimation that I could see myself being wrong (i.e. that a well-informed expert would take the other side of it).
Any idea how we can get in direct contact with one another without doxxing ourselves (e.g. posting our email addresses)?
I mean, it’s kind of both. His publicly presented timelines are usually insanely optimistic, but so far SpaceX has (eventually) accomplished most of what it has set out to do.
Falcon 9 is a great rocket now, but it’s major milestones all occurred years later than what Musk originally claimed (still historically fast). And Starship is way more ambitious. Rockets are hard.
I mean, Starlink is basically on track from the timelines he was talking about around 2015. You can go look at public statements and such, and they're remarkably accurate.
Falcon 9 was announced in October 2005 for a first half 2007 launch, actually flying in 2010. Falcon Heavy was unveiled in 2011 with a planned first launch in 2013, eventually not launching until 2018. Red Dragon was scheduled to launch to Mars in 2018, as publicly announced as late as 2016, then canceled entirely. Falcon 1, Falcon 5, 2nd stage recovery, propulsive landing for Dragon, and dry recovery of payload fairings were all canceled after slick Elon presentations proved to be infeasible.
Again, SpaceX has a remarkable record of major achievements. But not every idea has been successful (not even ones that get as far as major hardware tests) and very few have happened on Elon’s optimistic timelines. And Starship is more ambitious than anything else they have attempted.
60% chance of orbit this year is very very optimistic. My median guess for a successful orbital test of the system is second half of 2023.
While not disagreeing with the rest of your reply (I know nothing about SpaceX's history with deadlines, only Starlink's), is this really true? Proliforated LEO was still mostly a concept, not a reality back in 2015, and no one had ever mass-produced satellites in the way that Musk had proposed doing.
In hindsight, perhaps it looks less challenging, but I'd be interested to hear why, looking forwards in the early 2010s, Starlink appeared to be a much easier thing to execute than, e.g., Falcon 9.
Iridium did “proliferated LEO” in the 90s. Increasing the number of satellites by an order of magnitude or two is mostly a problem of affordable launch capacity (which Falcon 9 addressed), evolutionary developments in satellite technology, and the will to do it. Generally, doing lots of something you already know how to do is easier than doing something no one has done before - unknown unknowns are what get you.
It’s not that Starlink is necessarily easy, it’s that what Starship is trying to do is really really hard - the pure physics are much more challenging. Getting that many big rocket motors working in sync without blowing up is not at all trivial, let alone landing the thing on the same pad it took off from.
Propulsive landing for Dragon (after parachute, rather than sans parachute) wasn't human-rated by NASA, but as far as I know there wasn't any major unknown difficulty about it, as with catching fairings, which did turn out to be too difficult to be cost-effective for now after a number of attempts.
It wouldn't surprise me if propulsive landing for crew Dragon ended up significantly deprioritized partially due to the April 2019 failure. It was technically due to a different system, but I doubt there's much appetite to increase the volatility of crew capsules.
Then why also cancel Red Dragon, which was initially planned to be unmanned? I think some technical challenges e.g. Dragon getting heavier probably played a role as well.
Don't forget the original ITS timeline from 2016, with first unmanned Mars mission launching in 2022 and crewed in 2024. Not sure if Starship should inherit that one or if the whole system ought to just count as scrapped.
Posting predictions publicly to calibrate yourself makes sense. Not posting predictions about your and your friends' private lives publicly makes sense. The particular combination of those that you've chosen doesn't make sense.
Anyway good luck with Lorien. I'm very curious how that turns out, especially in this context since it's effectively the single biggest bet you've ever made. Really hope that it succeeds!
Maybe post the public predictions only? (Alternatively, I think it might be meaningful to precommit to the private predictions, then publish their SHA512 hashes or something along with their probabilities. Assuming SHA512 doesn't get cryptanalyzed into oblivion soon, this would allow one to keep private info private while still not leaving room to weasel out of past mistakes.)
I don't see how posting a hash is better than just posting "redacted" unless you plan to *eventually* make the thing public (allowing people to check that it matches the hash). Based on his grading of last year's predictions ( https://astralcodexten.substack.com/p/2020-predictions-calibration-results ) he doesn't appear to do that.
I have found that predicitng economics and finance stuff tends to be easy (GDP go up, stocks go up, rich ppl get richer, etc.) but outcomes of poltical events is way harder. This means you cna make easy $ by being bullish on stocks, most of the time, but prediction markets are more of a fool's errand.
Predictions that stocks will go up are right more often than predictions about political outcomes, but that doesn’t necessarily make them a better bet. It also matters what kind of odds you’re getting. With stocks, when they go down they tend to go down by a lot, so your implied odds are not 50/50. It’s like betting on a prediction market trading at 80c - if you bet “yes” you’re probably going to win, but that doesn’t mean it’s obviously a good trade.
I'm not sure I agree completely. It requires a deep knowledge of history, human psychology, and possibly the background of the individuals themselves, but - and especially in political games that rely a fair bit on populism - it is possible to be quite accurate.
Not to bray on my own behalf, but I predicted the mass protests we saw last year about a month in advance. I didn't know the political cause they would assume, but got the timing almost perfectly. We might counter that those were predictions of public sentiment and behaviour rather than pure politics, but that's exactly what opinion polls and the like are anyway.
I subscribe to the idea that Bitcoin has a strong correlation (70%+) to loose US monetary policy. There is no indication that the Federal Reserve will be more accommodative than it is today, quite the contrary, most likely by the end of the year there will be signaling of tightening monetary policy in the near future, negatively impacting Crypto and other ultra high growth assets.
there is no correlation tbh. That is just a media narrative to explain what is just speculation. Bitcoin had its best year even in 2013 despite the fed signaling tightening.
Probably a fair bit of the reasoning is simply that BTC and ETH have already gone up a lot, so it’s not unreasonable to bet that the trend will continue.
It seems like he just expects the bubble to be a little likelier than not to fail to burts by year end. I suspect that he’s wrong about that, but only because my fiancé’s idiot brother tried to get her to open a crypto-trading account that isn’t available in his state so she could make straw purchases for him. That has “bubble near burst” written all over it.
Scott, I’m half-trolling here, but do you have enough confidence in your BTC and ETH predictions that you’re willing to put money on them? I will take the other side of one or both, at whatever size you like. We can do the bet either for personal profit or for donation to a charity of the winner’s choice.
How much of your net worth is in BTC and ETH, because your predictions represent a level of bullishness that is much more bullish than the market. You can back out these percentages from options markets, deribit is currently implying a 15-20% chance ETH will be over 5000 by years end, you said 50% I would say that if you don't have the majority of your net worth in crypto your predictions are unreasonably bullish
Why would having the majority of his net worth in crypto make a prediction that it would go up more reasonable?
I can see an issue with "if you're more bullish than the market and confident, you should Go Full Kelly", but that's a consistency/confidence issue, not a reasonableness issue. If I predicted that everyone would abandon Facebook tomorrow and shorted several million worth of their stock, I'd be consistent and confident but not reasonable.
If you're going to put a significant amount of your net worth on the bet, it's also essential to factor in the likelihood that the value will tank. He doesn't make a prediction that the value will tank, so we don't have enough information to compute the reasonableness of doing that.
For example, maybe you think ETH will got o $5000 with 50% confidence, or will go to $1 with 30% confidence. Weird, but possible. Then, it's reasonable to put a small-to-moderate amount of money in, but probably a very bad idea to put the majority of your net worth in. Because if you put the majority of your net worth in, you're essentially going broke with 30% probability. You might think 30% of going broke vs 50% probability of getting obscenely rich is a great trade, but it's not, because money has diminishing returns. The the utility value of money is highly asymmetric.
All that to say, just based on the numbers we have, we don't know how good of an idea it would be for Scott to bet more on his BTC and ETH predictions.
You can back out a market implied pdf from options markets. Anytime your subjective beliefs disagree very strongly with this market implied distribution you can find a net that gives you incredibly good EV. I gave one example in my other post. At the time Scott posted this he could have bought an ETH call spread that would quintuple in value if eth reached 5000 by the end of the year. An investment with a 50% chance to quintuple in value is something you should be throwing lots of money at (37.5% of your bankroll by the kelly criterion).
Even if he doesn't want to mess with options and only bet through buying ETH he would have to have an incredibly unreasonable pdf on mind in order for it to both fit his predictions and not by an insanely good investment
Getting a degree in Elonomics so that I can understand the Elonomy was the best decision I ever made. Most media attention goes to the macro elonomy, but I find Microelonomics more interesting.
Scott has talked before about CO2 having effects on human cognition at levels low enough to be anthropogenic (as low as 800 ppm). So if you believe anthro climate change risk is only moderate due to geoengineering being able to arrest any tipping point effects (atmospheric SO2), and take a Hansonian view of human welfare, and believe that high CO2 levels could lead to e.g. IQ declines on the level of 5+ pts globally then actually that effect is potentially far, far worse for humanity measured as an integral.
Very excited for the essays mentioned at the end; what can I say, I love the lurid topics like climate change. Also appraisers are paid to find complications; 50% is too low. Also linking to/outing DSL is probably not a great idea.
I guess I'd say it has a tendency to indulge/skew toward the kind of direction Scott has tried to avoid his community going in, historically. cf More Right, The Motte, etc.
Your comment made me go search for the link (I'd previously assumed they were all just links to pages in his own blog) and follow it. An amusing mini-Streisand effect.
> 30. Some new variant where no existing vaccine is more than 50% effective: 40%
Recommend you clarify this. How certain does the evidence need to be? How prevalent the variant? Like, do you get this right if there's a variant that some case study finds 10 cases of and 6 of those people die? I think you technically would, but I think it would be pretty irrelevant.
Another way of thinking about it: I'm reasonably certain that this will be true for "some variant", so the question is really about whether it matters enough for anyone to notice. If you're really looking to predict something about the prevalence of a variant that is highly resistant to the vaccine, you should re-word/clarify.
Scott, I will take the other side of your Dow Jones predictions in unlimited size. The delta on the DIA 350 12/31/2021 call (roughly the probability of the DJIA ending above 35k) is 0.417 - you're way way off market.
Delta isn't the probability of an option ending up in the money. If you want to see what the market is implying you need to look at the price of a very tight call spread as this approximates a binary option
What do you consider to be BLM protests? Do any anti-police protests count, do they have to be endorsed by any particular group, or are you just counting any protest that is generally aligned with the goals of BLM?
Who defines clearly? "Scott Alexander thinks court packing is going to happen" or "The popular consensus (as assessed by Scott Alexander) thinks court packing is going to happen"?
I think this mostly comes down to whether they change the law that limits the size of the court. If they change the law to allow, say, 13 justices, then it's basically a sure thing that they're going to go ahead and appoint those extra justices.
Needs Manchin's approval, probably won't get it. Also needs filibuster abolition, right, since it's not budgetary? And Manchin has said no on that too.
Would be cool to put out these predictions first for a bit before showing your probabilities. I'd be interested in assigning my own without being biased by yours.
I don't see any major BLM riots. Police chiefs are making sure their officers aren't doing things that look horrible on video, and we now have a presidential administration that knows how to reduce tension in the black community, rather than inflame it. I see small scale protests whenever a black person gets killed by police, but nothing like what happened last year.
Many things, but off the top of my head he tweeted things like "when the looting starts, the shooting starts" and sent in Federal law enforcement against the wishes of local officials.
Had there already been major rioting by the time Trump tweeted that (dangerous and disgusting comment)? As a Trump-hater, I'd *like* to be able to assign him a decent share of riot-blame, but if they were already serious by the time he said that, seems wrong to blame him or predict low-intensity this time.
Nonetheless, my gut instinct is that Scott is a bit high on that one.
I mean, it's Portland – they're probably always going to be exceptionally likely to continue 'The Resistance' under any plausible Democratic Party presidential administration anyways.
I'm kind of impressed you view the odds of a major problem in the China/Taiwan issue at only 5%. Most watchers I've read are nervous about that, with the feeling that China is feeling its oats and the US has been somewhat prostrated by COVID and cultural conflicts, so...the time might be right. I'm a little curious whether you read deeply on the subject and decided that's all worrywartism, or whether this is more along the line of "well nothing's happened in the past 50 years of occasional concern so..."
I mean, IIRC Scott lives in the Bay Area (certainly on Moscow's list, and almost certainly on Beijing's and Pyongyang's), which kind of implies that prediction. If someone's bet their life on ~WWIII, it's kind of obvious that they don't think WWIII is very likely.
(You could get cynical here and say that given he's already betting his head there's no marginal risk in betting his internet reputation as well, but I don't think Scott's that mercenary.)
A fight between China and Taiwan won't necessarily involve the U.S., and even if the U.S. does get involved it won't necessarily escalate to nuking cities.
I mean, not *necessarily*, but it seems extremely likely in the conditional. I think after Hong Kong anybody who let Taiwan go without a fight would feel like Chamberlain, and while the USN is very mighty it's probably not up to fighting within range of PRC airfields. The US's ability to dictate terms, AIUI, consists pretty much solely of their ability to win a nuclear war via larger arsenal + allied close-range launch sites.
But what are the odds of an amphibious assault within the next 8 months? (really 6 due to unfavourable winter weather) They'd have to make the decision within the next few months if they want to be ready in time for the October window.
Over a year passed between the Trident Conference when the allies agreed on a cross-channel invasion and D-Day.
Well, not high, but I suppose I thought "major" event could fall short of a full-scale invasion and still qualify. To my mind "major" is probably anything that genuinely involves the USN. I'm pretty sure we (the US) could and would overlook any number of Taiwanese casualties that *didn't* imply an actual takeover, but if for example a US ship got hit that would be difficult to rationalize away, much as the current (or previous, honestly) Administration will want to do so.
I don't think a major event short of full-scale invasion is all that likely, since it's not in the interest of the PRC.
The 7th Fleet and ROCA are both kind of badly run. A major act of aggression from the mainland is likely to create the political will in Taipei and Washington to go in and fix these problems. It'd also make it easier for Taiwan to buy equipment like diesel-electric subs and modern air defense systems from other countries.
Taiwan is buying Patriot, wants to buy THAAD but that would be almost a red-line event for China, and is building their own modern air defense systems as well. Taiwan cannot buy diesel-electric subs from anyone because literally no nation capable of building diesel-electric subs is willing to sell arms to Taiwan.
What I mean is that if the PLA does something that is widely considered to be a dick move e.g. taking Kinmen, it'll shift the geopolitical landscape. Many countries will become more willing to sell arms to Taiwan and the big bad PRC will have less standing to complain about the ROC buying more modern systems.
In terms of realpolitik, a sudden aggressive move by China would anger a lot of countries at once and It'll be harder to push them all around with trade sanctions etc. than it would be to punish an individual country for getting too cozy with the ROC.
I've seen the mentions of PRC figures saying "we'll go in if Taiwan gets THAAD".
Can someone explain to me why they've got a problem with Taiwan having THAAD? They don't exactly need ballistic missiles to attack Taiwan - it's within range of bombers and of cruise missiles - and as it's not boost-phase it doesn't degrade their deterrent against the USA.
Their claim is that the THAAD radar is secretly meant to perform boost-phase tracking and battle management in a way that does degrade their deterrent against the USA. I'm skeptical that they really believe this, but they do say it when anyone tries to put a THAAD site anywhere near China.
I think it's much more likely they believe that yes, they do need ballistic missiles to attack Taiwan (or South Korea or Japan or anyone else). Or defend against an attack coming from those places. Yes, those places are within reach of China's bombers and cruise missiles, but those places are protected by air defenses which can shoot down China's bombers and cruise missiles. And China doesn't have Wild Weasels. Their edge comes from having lots of effective precision-guided ballistic missiles and their targets *not* having effective missile defenses.
You seem to assume that they haven't already made the decision. Is there some reason we would know if they had? The US military intelligence apparatus would probably at least suspect, but my understanding is that they don't publicise that sort of thing right away.
Some of the rhetoric at the Alaska summit (specifically, "Taiwan is a domestic issue", "we will have to take firm actions" and "if we confront each other, it will only damage the US; China will pull through") could be taken to mean "we are going to invade Taiwan; stay out of it or you'll lose".
Which of these are you suggesting is more likely? Both seem plausible but unlikely.
1) the Democrats have the ability to court-pack if they want, plausibly could run into a situation where it would benefit them (i.e. they're not missing a house or getting vetoed, but have an unfriendly SC and don't have the numbers to go the impeachment route), and publically threatened it after Amy Barrett was nominated. However, it seems relatively unlikely that they'd actually pull the trigger.
2) The petition to recall Newsom succeeded in getting the proposition on the ballot, but he's recalled if and only if the proposition passes. Current polling is at 60:40 or so in his favour from my digging, which is a large but not impossible gap to make up.
"[redacted] wins the book review contest: 60%" made me wonder. It seems unlikely that this is referring to one specific review, since it would have to stand out massively in a contest of some 20+ reviews (assumption based on "I’ll be posting about two of these a week for several months"). It could be a categorical statement like "A fiction review wins the book review contest", but that seems too easy. My prediction is that this is commentary on a fundamental issue with contests like these, for example how people are statistically more likely to choose whichever option was presented last as their favourite.
I agree that 60% seems high, but I would guess that is indeed for a specific review. Bear in mind that a) Only Scott has read all the reviews b) Scott probably knows already which review he's going to vote for, and c) Scott's choice will have (probably) the same weighting as everyone elses votes put together. Or something.
Or put it another way. If Scott granted only a third of the voting power to himself (on his own blog, running his own book review competition, and giving away lots of his own money) it would STILL be quite remarkable if the commentariat managed to disagree enough (and be unanimous enough) to outvote him. Probably also true if he only has a quarter of the votes.
In fact, 60% seems quite unconfidant...
Maybe the voting format is going to be different from my expectations.
>Somebody has to choose the winners, and it's either going to be me or the readership. I would rather people write things my readers like than things that I like, especially since I'm easier to manipulate (you all know what kinds of things press my various buttons).
You could interpret that various ways, but I think that means he's not going to reserve significant voting power to himself. He *did* already have a rather significant hand in choosing the winner (via picking the finalists), after all.
Fair point. I was vaguely remenbering some much earlier comment about how Scott and the commentariat would both have an input into deciding the winner(s). The comment you quote makes Scott's prediction much more understandable.
Aside from what Anteros said, it's also sometimes the case that things can be heavily favored with only a small difference in "quality" --
Think Biden being favored over Trump 80-20 by 538 in the 2020 Election, even though he was only a few percent better in the polls, or a sports team being heavily favored in a playoff series ever though their record is only 10% better. (Neither of these are exactly the same as the book review contest, of course)
Perhaps this review only stands out a little bit, but if that slight difference in quality is clear to a good percentage of voters, that could be enough to make it strongly favored.
I have added these to a foretold community (https://www.foretold.io/c/6eebf79b-4b6f-487b-a6a5-748d82524637), and predicted on the non-redacted ones. It might be a good place for the commentariat to make their own predictions. Making predictions requires an account, but viewing them doesn't.
Because the number of questions is relatively large, (75 non-redacted ones), it would be relatively simple to have a tournament, with a relatively simple scoring rule/winning criterion.
Looking forward to the reviews of Nixonland and How Asia Works. Not to snub The Scout Mindset, it was a valuable book that I'm more likely to recommend to friends than the previous two, just not as excited to see your take on it.
You can meditate with a traditional mindfulness technique for 3 minutes a day effectively and have a far greater accumulative impact than doing 10 vippasina retreats or squirming in fits of boredom alone with an app never knowing if you’re making progress. More importantly - Direct transmission from a realized lineage holder/teacher > apps or secular modern neuroscience based meditation and its rigid container. Exponentially more profound realizations of time/form/phenomena/mind/afflictive emotions vs simplistic stabilization is avail w a lineage teacher, group, setting. The focus through breath (Annapurna) or even impermanence insight meditation is scrunched into discursive mind without deeper containers.
Or.. one can just sit and do Metta/emptiness meditation properly with open heart. Even for a minute. Actual liberation is experienced rather than parlour games of dilating focus through a misapprehension of mind and causal interdependence. Just saying! :)
Bleh – the main thing that people that are a "lineage holder/teacher" ('guru') seem to be good at is being (or pretending to be) an 'impressive lineage holder'.
The people that actually seem to be impressive or successful, or even 'just' _enlightened_ – in my experience anyways – seem to strongly discount the potential impact of meditation.
(I also suspect that almost everyone regularly engages in something that's functionally equivalent to meditation.)
Most of the supposed benefits – which are _traditionally_ all self-reported – seem more likely to be the placebo effect (i.e. wishful thinking).
And any honest (and accurate) accounting of the overall effects of meditation practice needs to grapple with The Dark Side of Meditation. I've never seen anyone try to reasonably offset that and even roughly determine whether meditation is good or bad on net.
Bleh – the main thing that people that are a "lineage holder/teacher" ('guru') seem to be good at is being (or pretending to be) an 'impressive lineage holder'.
---You've not shopped around sufficiently. Or your'e parroting an anti teacher trope favoured by the western degradadation of Buddhism and contemplative work in this era. Yes, egos abound with teacher types. But... egos abound period (and to all types of degrees and qualities.)---
The people that actually seem to be impressive or successful, or even 'just' _enlightened_ – in my experience anyways – seem to strongly discount the potential impact of meditation.
--That's curious. Unless someone is realized at birth, or through traumatic intervention or an amazing confluence of events resulting in the ability to directly realize mind/interdependence and emptiness etc etc... I think you're confusing enlightened with..morally upstanding or wise. They are not separate but neither are they always (even often) the same.--
(I also suspect that almost everyone regularly engages in something that's functionally equivalent to meditation.)
Most of the supposed benefits – which are _traditionally_ all self-reported – seem more likely to be the placebo effect (i.e. wishful thinking).
Yes, we should jettison 5k years of vedic and western contemplation and texhniques, profound cannonized teachings based off of them and the direct expression of millions of humans today. All just delusional yearning into the void and conformation biasing some deeper confusion? No.----
And any honest (and accurate) accounting of the overall effects of meditation practice needs to grapple with The Dark Side of Meditation. I've never seen anyone try to reasonably offset that and even roughly determine whether meditation is good or bad on net.
-- You/one will experience the shadow on the way to the light. We are not after happiness as it is understood through the sensual lens. That is the obstacle. We are after liberation and it includes a very dangerous path of recognizing how corrupted our perception and reasoning is. That is what one discovers. Yes, it is traumatic. And beyond that resides a joy. A true joy. I ain't trying to sell you anything. :)
> You've not shopped around sufficiently. Or your'e parroting an anti teacher trope favoured by the western degradadation of Buddhism and contemplative work in this era.
I would imagine that you're so sure of this because you're enlightened, huh?
> I think you're confusing enlightened with..morally upstanding or wise. They are not separate but neither are they always (even often) the same.
What's the point of enlightenment if it doesn't also mean (at least) _more_ moral or wise? The Western movement with the same name was, and I think 'properly', very much concerned with both morality and wisdom (even if, in practice, it fell short of either, or both).
> Yes, we should jettison 5k years of vedic and western contemplation and texhniques, profound cannonized teachings based off of them and the direct expression of millions of humans today. All just delusional yearning into the void and conformation biasing some deeper confusion? No.
I didn't write or even imply your summary.
I do think a _lot_ of it is something like "delusional yearning into the void and conformation biasing some deeper confusion".
But it's a big topic/subject/area-of-experience! There's, as you point out, _thousands_ of years of activity around this. There definitely _is_ real, significant, insight, wisdom, and even ethical and moral efficacy, tho sadly it's, like all traditions, 'protected' as sacred without being understood.
> You/one will experience the shadow on the way to the light. We are not after happiness as it is understood through the sensual lens. That is the obstacle. We are after liberation and it includes a very dangerous path of recognizing how corrupted our perception and reasoning is. That is what one discovers. Yes, it is traumatic.
I mean sure – I appreciate just the _idea_ of things like 'charnel ground practice'.
And I don't think "happiness" is a proper ultimate aim (or 'terminal value') either.
But I also don't think 'happiness' is an "obstacle", let alone _the_ obstacle, and I think aiming for "liberation" is, if anything, WAY more stupid (and evil) than aiming for 'happiness'. 'Liberation' in a metaphysical/mystical/spiritual sense can only mean one thing – death. A truly enlightened nihilist would promptly die of dehydration (not that the actual death would arrive promptly).
Some amount of liberation is perfectly fine!
I don't think existence is something from which beings need to be generally 'liberated'.
And you write "how corrupted our perception and reasoning is" whereas I think that's silly and obviously 'charnel ground practice'. I think the boring/mundane truth is just that ANY perception or reasoning, by any kind of 'being', must be boringly/mundanely 'implemented' _some_ way, and _one way of looking at_ that brute (and Sad) fact is that any perception or reasoning is necessarily "corrupted". I think 'limited' is a more enlightened perspective.
I don't think The Truth is traumatic either – it sure is for some people tho, I'll admit. But maybe that's just because we're all generally confused about what it is, or how best to reach it (or whatever peak(s) we've already discovered).
> And beyond that resides a joy. A true joy. I ain't trying to sell you anything. :)
"A true joy" huh? <sarcasm>Yeah, I have NO reason to suspect you of selling me a wireheadseat.</sarcasm>
> Major flare-up (significantly worse than anything in past 50 years) in China/Taiwan conflict: 5%
Just a reminder that the past 50 years include:
- the PRC and ROC militaries exchanging symbolic artillery fire with propaganda leaflets on alternate days during the 1970s
- the 1987 Lieyu massacre, during which 19 Vietnamese boat people were killed by the ROC military after stranding on a militarized island near Kinmen. (Kinmen and surrounding islands have since been put under civil administration starting in the 1990s.)
- the Third Taiwan Strait Crisis (1995-1996) when the PRC military fired ballistic missiles into the waters near ROC-controlled islands, including Taiwan.
- PRC military aircraft crossing the center line of the Taiwan strait in 2021, but at an angle so they were not flying directly towards Taiwan.
My guess is that when establishing the baseline of "significantly worse than anything in past 50 years," you were thinking about some, but not all of these events.
My alternative predictions:
- PRC military aircraft cross the Taiwan strait at least two thirds of the way, on a course headed straight towards Taiwan: 5%
- PRC military conducts live-fire exercises in waters near ROC-controlled islands in a repeat of the Third Taiwan Strait Crisis: 2%
- PRC artillery hits land controlled by the ROC: 1%
- civilians are killed by troops fighting in the Taiwan Strait: 0.5%
- PRC gains control over Taipei: 0.1%
- PRC, ROC and US militaries continue to make plans involving a PRC invasion of Taiwan as a possible scenario: 99%
I know very little about the specific details of the China/Taiwan situation besides the knowledge that comes from reading mainstream news articles, plus the book On China. But I do think that when making probabilistic forecasts of any kind, it’s very difficult to justify estimates less than a few percent. I think the world is just generally far more unpredictable than people think. So on that basis, I would argue that your probabilities should not tail off as aggressively - maybe they should ramp from 5% down to 1% or 0.5% rather than 0.1% (with intermediate scenarios adjusted proportionally).
0.1% is easily reached by a combination of three factors with 10% chance each. It's probably not very helpful to include events that unlikely when making only a few predictions, though.
For the PRC to annex Taiwan, it'd have to invade and the USA would have to decide to allow it (the PLA is not sufficiently powerful and decentralised to eat several hundred nukes and then mount a contested amphibious invasion against 1.7 million Taiwanese reserves - outright invasion of China is probably beyond immediate US capabilities, but stopping a boots-on-the-ground invasion isn't). Predicting 2% invasion/1% PRC takes Taipei is implicitly predicting 50% conditional of "USA allows Taiwan to be invaded", which seems way too high (it's not like it's a domestically contentious issue; the one major thing Tumblrites and Trumpists agree on is "Beijing bad", and allowing a US ally to fall uncontested would be permanent political and legacy suicide).
I agree. I think an estimate around 1% is more reasonable than either 5% or 0.1%.
Integrated over all "plausible conflicts" (which are maybe in the order of tens) it would mean that there is one conflict every few years that breaks out without it being majorly foreshadowed, which sounds about right. Although I am a little skeptical of that line of reasoning, as state-vs-state conflicts nowadays have higher hurdles, and if you only count state-vs-state scenarios you have much different base numbers.
made the mistake of thinking Scott was prediction an approval rating for Biden of between 50% and 80%. Fortunately, I caught it before I posted to say Biden wasn't going to get an 80% approval rating, but I do think the prediction was easy to misread. It needs more space between the two percentages or something. Or prediction percentages should be a different color.
Also, I don't see the point of giving each redacted prediction its own line. If you want to mention them, how about just saying how many of them there are in each category?
The FDA doesn't like approving things. Even now when I got the vaccine I had to read an FDA statement saying that they were not approved, just allowed because of an emergency. So I think betting that the US won't approve a vaccine is generally a good idea.
The United States doesn't need the AstraZenica vaccine for its own purposes, and it takes substantial political capital to make the FDA approve anything in less than years. So the question is, how much political capital is the Biden administration willing to spend to be able to donate vaccine to India? Not zero, certainly, but also not certainly enough to get it done.
I think Scott does sometimes throw out predictions if the result is hard to call one way or another. Hopefully this will be one of those where there will be basically no conflict and it's easy to judge...
> 21. Google widely allows remote work, no questions asked: 20%
You probably mean specifically fully time employees who are knowledge workers at Google, correct? In this case it might be useful to add more contextual qualifiers to your predictions at the risk of making the list a bit wordier.
Am I insane or was there not already a review of Nixonland published here? I just looked in the archives, though, and this post was the only one that came up.
63. reminds me that the US is still a dystopia that accepts paper walls, flimsy houses in general and incessant TV consumption as completely normal.
Thank god for the 50+cm of solid stone all around me... No AC necessary, either!
I've recently learned about Heat Pumps and their incredible efficiency at heating and cooling no matter the surrounding temperature. Although I couldn't find numbers for how efficient they are considering the transport waste of electricity... Promising nonetheless!
Your comment reminded me of G. K. Chesterton's shock when he visited the American countryside and found that all our houses were made of wood! Before I read that account I never thought there was anything particularly remarkable about a wood frame house.
"The sharpest pleasure of a traveller is in finding the things which he did not expect, but which he might have expected to expect. I mean the things that are at once so strange and so obvious that they must have been noticed, yet somehow they have not been noted...I had no sooner passed out into the suburbs of New York on the way to Boston than I began to see something else quite contrary and far more curious. I saw forests upon forests of small houses stretching away to the horizon as literal forests do; villages and towns and cities. And they were, in another sense, literally like forests. They were all made of wood. It was almost as fantastic to an English eye as if they had been all made of cardboard. I had long outlived the silly old joke that referred to Americans as if they all lived in the backwoods. But, in a sense, if they do not live in the woods, they are not yet out of the wood.
…
These things seem somehow to escape the irony of time by not even challenging it; they are too temporary even to be merely temporal. These people are not building tombs; they are not, as in the fine image of Mrs. Meynell's poem, merely building ruins. It is not easy to imagine the ruins of a doll's house; and that is why a doll's house is an everlasting habitation. How far it promises a political permanence is a matter for further discussion; I am only describing the mood of discovery; in which all these cottages built of lath, like the palaces of a pantomime, really seemed coloured like the clouds of morning; which are both fugitive and eternal.
…
I was always told that Americans were harsh, hustling, rather rude and perhaps vulgar; but they were very practical and the future belonged to them. I confess I felt a fine shade of difference; I liked the Americans; I thought they were sympathetic, imaginative, and full of fine enthusiasms; the one thing I could not always feel clear about was their future. I believe they were happier in their frame-houses than most people in most houses; having democracy, good education, and a hobby of work; the one doubt that did float across me was something like, 'Will all this be here at all in two hundred years?' That was the first impression produced by the wooden houses that seemed like the waggons of gipsies; it is a serious impression, but there is an answer to it. It is an answer that opens on the traveller more and more as he goes westward, and finds the little towns dotted about the vast central prairies. And the answer is agriculture. Wooden houses may or may not last; but farms will last; and farming will always last.
6. Significant capital gains tax hike (above 30% for highest bracket): 20%
Seems too low, due to democratic control of everything at the federal level, plus corporate lobbyists not really caring one way or the other about individual tax rates, plus an unprecedented deficit and debt/gdp ratio and rising interest rates on US treasury bonds for the last ~8 months since the bitcoin bull market began (negative real interest rate bonds are a sucker bet that will lose market share to crypto)
7. Trump is allowed back on Twitter: 20%
Seems waaaay too high. The prior is that twitter almost never unbans anyone (especially not months later -- I haven't seen a single instance of that -- errors are corrected within a couple weeks or never). Plus the social dynamics inside twitter would make it even harder to unban him. It's full of woke employees, so championing the cause of free expression as applied to that particular case would be career suicide there. The democrat controlled congress certainly isn't going to pressure twitter to unban trump either. I'd say 1%.
The likelihood of #29 is probably higher than 50% on a standalone basis.
But 40% for #30 almost requires a higher % for #29: If a new variant evolves for which current vaccines are less than 50% effective, it will surely be very widespread relative to existing variants and therefore likely responsible for >25% of cases.
The two exceptions are if there are multiple variants like this (therefore preventing either from claiming a large share on its own) or if such a variant appears very late in the year, thereby fulfilling #30 but not having enough time to reach a large share.
the DEC31 3840-5000 call spread is offered at 240 on deribit right now, that spread is worth strictly more than a binary option that pays out 1160 if this prediction is true. That is to say you are giving something a 50% probability that you can bet on at 4:1 odds for large size. That kind of disagreement with the market is bold, not necessarily wrong but bold. If you stand by this prediction in light of how stongly the market agrees you should probably try to get set up to trade crypto options, its a pain as a US citizen but not impossible.
> Fewer than 10K daily average official COVID cases in US in December 2021: 30%
Ouch, that's within a factor of 2 of where we are right now, and 70% chance we'll be worse than that. So you're really thinking this isn't going away anytime soon. I was hoping we'd see it dwindle to where we could all ignore it, especially for the sake of the kids too young to get vaccinated (and who are unlikely to die, but might get damaged in poorly understood ways).
I'd be interested to hear more about how you think the next year is going to play out. (Especially combined with your pessimism about vaccine escape).
"Court packing is clearly going to happen (new justices don't have to be appointed by end of year): 5%"
I assume that this simply means "Court packing is ... going to happen: 5%." Otherwise, I'm confused about what the 'clearly' means here. Is there then a separate prediction for it *not* clearly happening?
I think the point here is that court packing hasn't yet happened by 2021 end, but decisions have been made, as of 2021 end, to pack the court after 2021 end.
Right; even if the Democrats manage to make "court packing" happen, there's likely going to be a delay while Democratic progressives try to pack it with maximally-liberal justices and Joe Manchin says "that's not what I signed on for; try again". Or maybe Biden is a gracious winner and only nominates respected center-left judges for quick confirmation, but it's reasonable for Scott to give himself some wiggle room on that.
> 105. I finish and post the CO2 essay I'm working on
I would like to humbly suggest that you might find some useful information about CO2 on the blog Wonky Thoughts by Doug Robbins. He writes infrequently, but the posts contain a lot of data.
>If Ethereum or Bitcoin had a 40%-50% chance of doubling or better by end of year, every investor in the world would be putting all their money in it.
Not if they thought it had an equal chance of losing more than half it's value. If you see crypto as still somewhat in its early stage of price discovery where the supply is very well known but the demand could potentially increase a lot if more mainstream investors get involved (especially US retail investors once some bitcoin-based ETFs become available) then there's a good chance the price could bubble up past 100k. There's too many unknowns to say EMH should apply just because it's gone a bit mainstream.
>Not if they thought it had an equal chance of losing more than half it's value
Shouldn't this be losing all its value? If I can buy something for 100 and there's a 50% chance it will be worth 200 or 50 in a year, that's an ev of 125, and a good deal (ok, not for *all* my money, but there should be enough money around that would take an opportunity like that)
Jack's claim was "would be putting all their money in it". Kelly betting says you bet to maximise E(log(wealth)); when you're betting all your money, therefore, 50% of double/50% of halved is exactly even (50% of double/50% of -100% has E(log(wealth) = -infinity; you should never make a bet with a nonzero probability of total ruin, since then you can't make any more bets).
Kelly betting approximates maximising E(wealth) if you're not betting very much of your net worth (linear approximation), but you need to consider the nonlinearities when "all their money" is invoked.
Inflation is still low though according to the data such s CPI and barely budged despite all this printing. The expansion of the money supply is mostly going to assets and paying off debts or saved, not being spent on consumer goods as much or in such a way that it would create inflation in terms of rising price levels. Even if consumer spending does go up, this does not mean inflation if companies can increase production to meet demand and not raise prices.
How would someone manipulate CPI calculations? Isn't this all based on public information, with a pre-specified algorithm?
If you look at the links provided, that's all done by mathematical calculations: https://www.bls.gov/cpi/quality-adjustment/home.htm
https://www.bls.gov/cpi/quality-adjustment/questions-and-answers.htm
I suppose the people who are out there doing research on men's apparel might choose the brands to sample in such a way that the "hedonic adjustment" calculation moves in one direction for a few months and then the other direction for a few months, but they're not going to be able to do anything systematic for an extended period, and they'd have to somehow collude across many dozens of categories to even produce a systematic short-term effect.
The full quote from the remote work article is "Employees can apply for up to 12 months but only those deemed to be in “exceptional circumstances,” will have their request fulfilled.", which says the opposite of what you're saying it says.
Firstly, how do you know that? Secondly, even that's a very different standard from Scott's prediction criteria of "widely allows remote work, no questions asked".
Still it sounds like a question is asked.
As a counter point to the above (commenting here for the Google wfh point)
1) his 80% is the probability of Biden having a higher than 50% approval rating. That seems quite plausible to me.
2) I work at Google and my reading of that question is that anyone who wants to could work from home, full time, even outside of their office city. I think Google management has been pretty explicit / public about not committing to that - Id rate the probability <5%. Which doesn't mean that some people on the margin won't be able to work remote / not come back for an extended period by negotiating with managers, just that this won't be the norm.
He's not saying Biden has, or will have, an approval rating of 80%. He's saying that there's an 80% chance of Biden having an approval rating of 50% or better on Jan. 1 2022.
The wokies don't like Biden, that's true. Non-woke Democrats do like him (that's how he won the primary), and swing voters like him (that's how he won the general). He's popular with regular people who don't yell at each on Twitter.
If you want to know the metric Scott is tracing, it's currently at 54.5%, and has been at 53.0% or above for all but three days of Biden's presidency: https://projects.fivethirtyeight.com/biden-approval-rating/
How is that? The election results don't show anything about polling - just about the difference between the polled population and the voting population. "Approval rating" isn't about voting though - it is just *defined* as the results of these polls, so there's literally no way for polling error to exist.
Anyway, the election results were off by less than 5 points everywhere, and the current approval rating shows a 14 point margin, so I think that anyone who has properly taken the lessons of the past few elections should accept that Biden's positive favorability rating is "real", whatever that means.
If it's defined as the results of those polls, it's sort of a problem, right? Phone polls recently seem to have a response rate of something near 1%. Approval rating that's the result of polling the self-selected 1% of the population seems like a meaningless number.
This is also why the election results were so wildly off. If you just take that 1% and try to weigh for all demographic factors, you can't assign enough weight to the group of people who just won't answer the phone poll, because they are not in your data at all! (I know there was at least one pollster who tried, though, by asking questions like: "do you think your neighbors will vote for Trump?")
Fair enough, but the prediction in question here is specifically regarding the 538 polling numbers.
So how do *you* measure Biden's approval rating if not by looking at polls?
Q drops, obviously.
Biden's approval rating is currently 54.3%
Kamala will be president before the 2024 election, not counting episodes as "acting" president: 5%.
That seems far too low. As a 78 year old man, Biden has a 4.7% chance of dying just in the next year according to this table https://www.ssa.gov/oact/STATS/table4c6.html ; we'd need to adjust that for his apparently-good health, but I doubt he's got a 95% chance of living to November 2024.
What are you expecting as cause of death?
Hmm... stacking the probabilities over the next four years, it's a little over 20%... and adjusting it a little downward, so about 15%? I suppose I was underestimating it, because I've never heard "Biden is likely to die" from anyone who isn't obviously in my outgroup, like Amy here.
In addition to his apparent health, you have to take into account the fact that he has regular access to the best medical care that money can buy. Presumably, he will be able to intervene very early and aggressively against any health issue that might arise.
Well, many people were happy to forecast that Trump would keel over dead some time in office, as he was so unhealthy, old, and ate a poor diet, but he's still alive. I am willing to assume that Biden may manage to live another four years to the age of 82 in November 2024. There has been some speculation about his cognitive health but there is nothing concrete about his physical health.
Possibly more likely is that in a year or two he may resign due to those cognitive difficulties to hand over to Kamala, but prognosticating his death does seem a little premature.
He didn't say Biden would die, he said it seemed like more than a 5% chance, which seems obviously true going by those tables, unless you think he's three times healthier than the average septuagenerian.
""Biden approval rating: 80% ???"
Pretty sure he means likelihood that Biden's approval rating at the end of the year will >= 50% is 80%.
Note that I'm not predicting Biden's approval rating will be 80%, I'm saying there's an 80% chance Biden's approval rating will be above 50%. Currently it's 54%.
Get his ass, Amy. Don't let Biased Nate Pig Iron and his made-in-China graphing calculator skew the polls!
"Independent but the Left is worse" and "The Bay Area is a hellish sea of NIMBY hypocrisy" are two viewpoints that I rarely get exposed to on these polyonymous codices! You may consider my mind duly blown
This is what “independent” right wingers always say. Just before they consistently vote for GOP candidates at every level of the ticket.
weird - 538's prediction was pretty much dead on
Also....538 didn't actually _conduct_ any polls, nor are they a polling company. This person clearly has no idea what they are talking about.
why I referenced their prediction - not the non-existent polls
Yeah, sorry, in case it wasn't clear, I was agreeing with you.
The fact that you think 538 conducts polls indicates that you are uninformed enough on this topic that I can safely disregard your opinion. You should probably bother to read some of their reporting on the topic before so widely and confidently expressing your opinions on the matter.
How will this resolve if Biden is dead at 31/12/21? Postmortem approval rating, last premortem approval rating, or "inconclusive"?
Probably that is a condition under which Biden does not have approval rating over 50%.
"- Inflation rate significantly will go up: 100%"
100% confidence is never correct.
Src: Bayes Rule
Redacted?
From the intro: "Some predictions about my personal life, or that refer to the personal lives of other people, have been redacted to protect their privacy."
"Vitamin D is generally recognized (eg NICE, UpToDate) as effective COVID treatment: 70%"
Isn't this counter to what you previously said on Vitamin D/COVID? Have you changed your mind, or do you expect the consensus to come to an incorrect conclusion? Or just a typo?
I'm skeptical that Vitamin D "really" works, but wouldn't at all be surprised if it gets recommended anyway.
I'm also curious about this, and if you've changed your mind about it, I think it would be worth adding a note to that effect to the original post (or if you don't have time I can at least add a comment there).
Yes, I forgot to add a "not". My mistake!
>25. Greater than 66% of US population vaccinated against COVID: 50%
50% seems too low to me, what's your reasoning here? Already more than 50% of adults have received at least one dose, and it looks like vaccines will be approved for children too soon.
I would put it at 80%
There will be a sorta trail off at some point as you start to hit the subset of people who don't want to get vaccinated for various reasons.
Notably, if you look at Israel, which has led the curve of vaccination, it seems to have trailed off around 60%: https://ourworldindata.org/covid-vaccinations . The UK also seems to be leveling off.
In addition, there's been polling showing as much as 27% of the US general population are hesitant to get the vaccine: https://www.kff.org/coronavirus-covid-19/report/kff-covid-19-vaccine-monitor-december-2020/
If covid starts to reduce its infection rate enough, and enough regulations are relaxed, I won't be surprised if that reduces the number of people who want to get vaccinated further, and there's a definite chance of news events, akin to the J&J blood clots, but even more sensationalized, which impact the general public's attitude more.
I don't think Scott's prediction is too bad given how much Israel and the UK appear to be leveling off in their vaccination rate. I'd personally give it more than 50%, but I think 50% is defensible.
Are you including "Medically contraindicated" in your 27%? I Suspect that number is <1%, but I've not seen any data beyond a few anecdotal instances.
Israel's numbers are complicated by the fact that children are not yet eligible to receive the vaccine. If you look at what percentage of the population that _is_ eligible has received it, I think you'll find something closer to 90%.
Which is why I'd like to see Scott clarify that prediction: is that of the entire US population, or of the subset of the population that is eligible to receive the vaccine, as 12/31/2021?
The UK is levelling off because the people who got their first dose in Q1 now need a second dose, so nearly all the supply is being allocated to second doses, with very little left over for new first doses. There are still plenty of people under 45 who aren't eligible for a first dose yet but really want one.
Yeah, switching the chart to "Vaccine doses" rather than "People vaccinated" doesn't show any leveling off, which suggests its down to devoting capacity to second doses. Anecdotally, I got my vaccine this week and the center was running pretty much at full capacity, so I don't think demand is the limiting factor yet.
Israel does show a leveling off for total doses too, which does seem to be due to lack of people applying, but as others mentioned a lot of this is due to children not being vaccinated yet: around a third of Israelis are under 16, so a 60% vaccination rate is actually most of those eligible.
I was thinking the same thing at first, but then realized that the adults/children thing causes extra complications. The NYTimes is saying that 80% of Americans are 16 or over. While I think they're likely to approve vaccines for ages 12-16 in a few months, it seems less clear that they'll approve them for younger people, so there's a reasonable chance that only 85% of Americans are eligible before December. So the question is basically whether more than ~80% of eligible Americans will eventually get vaccinated.
Since we're having a major slowdown in the adult vaccination at 50%, it's definitely not obvious that we'll get to 80% by the end of the year.
Maybe a lot of it really is the inconvenience and hassle for some people and general unawareness for others, rather than specific unwillingness, and maybe they'll gradually develop "get-out-the-vote" tools that work for these groups.
Also, maybe Novavax will get approved, and the people who are worried about adenovirus/mRNA "reprogramming their genes" will get that shot. (I actually heard someone specifically say he was waiting for Novavax for that reason!)
Yeah, I thought we were already past that for the adult population.
According to a recent poll, only 69% of Americans say they want to get vaccinated. That number could go up as people learn the vaccine is safe. But it could also go down, as COVID gets better and people stop feeling it's as urgent.
Individuals likely change their minds back and forth a lot in ways that you can't see as % changes on those surveys (because they mostly cancel each other out). There only has to be a brief window where a person approves of the vaccine in order to go out and get it and then they're vaccinated forever even if they change their mind again. But also most consumer research is bunk because people don't actually do what they say they're going to do. Maybe people answer those surveys according to perceived tribal stances but when their own skin is on the line they act differently. But the fact that 50% already got it makes me want to assign >50% probability to 66%.
What is the biological function of oroxylum?
Upvote. I googled expecting some ready links to the bioohacker's guide to oroxylum and...bupkis.
I found a plant that's used in Ayurveda medicine.
Oroxylum indicum's active ingredient, oroxylin A [1], appears to be a relatively understudied stimulant of the dopamine reuptake inhibiting variety [2].
[1] https://en.wikipedia.org/wiki/Oroxylin_A
[2] https://link.springer.com/article/10.1007%2Fs12272-013-0009-6
I had to look it up, and it has several names including such soft and emollient ones as MIDNIGHT HORROR and BROKEN BONES 😀
It's used in traditional medicine, and this paper doing a round-up on it indicates that it will do everything short of ironing your shirts and cooking your dinner:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3425058/
"Various parts of the plant are used in Ayurveda and folk medicine for the treatment of different ailments such as cancer, diarrhea, fever, ulcer and jaundice. Recent in vivo and in vitro studies have indicated its anti-inflammatory, antiulcer, hepatoprotective, anticancer, antioxidant, photocytotoxic, antiproliferative, antiarthritic, antimicrobial, antimutagenic and immunostimulant properties.
Roots are sweet, astringent, bitter, acrid, refrigerant, anti-inflammatory, anodyne, aphrodisiac, expectorant, appetizer, carminative, digestive, anthelmintic, constipating, diaphoretic, diuretic, antiarthritic, antidiabetic and febrifuges. Tonic is useful in dropsy, cough, sprains neuralgia, hiccough, asthma, bronchitis, anorexia, dyspepsia, flatulence, colic, diarrhea, dysentery, strangury, gout, vomiting, leucoderma, wounds, rheumatoid arthritis and fever. Root bark is used in stomatitis, nasopharyngeal cancer and tuberculosis. Leaves are used as stomachic, carminative and flatulent. Leaf decoction is given in treating rheumatic pain, enlarged spleen, ulcer, cough, and bronchitis. Mature Fruits are acrid, sweet, anthelmintic, and stomachic. They are useful in pharyngodynia, cardiac disorders, gastropathy, bronchitis, haemorrhoids, cough, piles, jaundice, dyspepsia, smallpox, leucoderma and cholera. Seeds are used as purgative. Dried seed powder is used by women to induce conception. Seeds yield non-drying oil used in perfume industry. The seeds are ground with fire soot and the paste is applied to the neck for quick relief of tonsil pain. The seeds are used in traditional Indian Ayurvedic medicine, included in famous tonic formulations such as Chyawanprash. Bark decoction is taken for curing gastric ulcer and a paste made of the bark powder is applied for mouth cancer, scabies and other skin diseases. The medicated oil of O. indicum in sesame oil base instilled into ears mitigates the pain in otitis."
See, this is why every time I try to learn about medicinal plants I run into a roadblock. Every plant I research supposedly does 12 different things! And every single one is a diuretic!
Finding out that every plant is a diuretic really p***es me off.
This probably means that it does nothing.
Scott should be in the correct sort of climate to grow Oroxylum indicum, right? https://www.chilternseeds.co.uk/item_926A_oroxylum_indicum
Really looking forward to the Oroxylum self-reports and the culture wars essay! I predict that your scores for these predictions will surpass all previous years.
You absotively-posilutely need to specify some kind of method or measure for 5. in advance, because it's going to depend a lot on (i) woolly estimation of numbers (ii) judgments about what constitutes "BLM" (iii) needing a dataset for this with a track record of year-to-year regular methodology
I think it could be an effective prediction as it is on the basis of damage *claims* of $250 million due to 'BLM protests,' rather than a specific pinning-down of what would constitute such damage.
Most news articles about last year's protests said "caused $X in damages". I'll judge this correct if a trustworthy news source says there were protests that caused $250 million in damage, unless some other news source says they're wrong, in which case I'll use my best judgment.
Is the prediction for the whole year or does it exclude those that have already occurred?
What total are you using for last year's?
you need to define what you mean by "vaccine effectiveness." for instance, my daughter still got mild chicken pox, localized to her calf and a one-day fever, even though she was vaccinated. does this mean the vaccine wasn't effective?
Yes, exactly. If your implicit definition in #30 is "variant makes vaccines <=50% effective at preventing severe illness and/or death in fully vaccinated people" then 40% probability of that happening this year seems way too high-- if that is your definition I would really be interested to understand how you arrived at that probability.
Have you written up why you've somewhat changed your mind on Vitamin D?
I forgot to add the word "not" in this prediction! Awkward!
>I've read and reviewed at least two more dictator books
I'm sure you already have plenty of options, but you might find Professor Archie Brown's book The Myth of the Strong Leader interesting in this regard:
https://www.amazon.com/dp/B06XKQ52TV/ref=dp-kindle-redirect?_encoding=UTF8&btkr=1
I haven't yet read it myself, but I greatly enjoyed and would highly recommend Prof. Brown's books The Rise and Fall of Communism (now on sale on Kindle for $2!) and The Human Factor: Gorbachev, Reagan, Thatcher, and the End of the Cold War.
Not a lot of high odds on those political predictions. I don't blame you - 2021 feels like a weird year, and that's saying something given 2020.
I just don't know with Yang, since NYC has ranked choice balloting. He could lose from that I think.
Really hoping #92 pans out! I'd love to read a full review of "How Asia Works". Very interesting book.
Yeah, I plan to put Yang at or near the bottom of my ranking just based on his platform still containing Basic Income garbage. Not sure how many other New Yorkers see it my way but I'd wager it's not going to help him.
What are your thoughts on why Basic Income is garbage?
I don't know what to feel about Yang's chances. I'd love to see him get the job, if only to see how his bright optimistic technocratic policies ram head-on into the wall of New York bureaucracy and institutional inertia, not to say entrenched and rampant corruption and crony politics. Sorry, Andrew!
On the other hand, there are other candidates I feel would do better - Stringer and Adams, for two. I was rather surprised that the news stories were all about the Democratic candidates, but it seems like the Republicans couldn't even catch a cold in NY governmental politics and non-party/independents are just novelty candidates with no hope at all. So it's going to be a Democrat getting elected. Yang is the frontrunner, but I dunno - I always have a feeling that people who were part of the bureaucracy/local government previously have an advantage, hence why Stringer and Adams (and maybe Garcia and Donovan, although they're both low in the polls, and I don't know if being the former counsel to de Blasio is going to help or hurt Wiley) seem to me to have a better chance when it goes to the polls with the ordinary voters, especially if they're using ranked choice voting this time.
So Yang could win the Democratic nomination but with five candidates on the ballot, the public might decide "hey, that guy OtherCandidate is the one who sorted out my problem with ApplicationForm" and go for them as Number One choice instead.
I said this last year too, but you’re way too bullish on Starship. Starship might reach space this year, but reaching ORBIT requires a fully functional Super Heavy booster, which is on version 1 “production pathfinder” right now (Starship itself is on version 15, and has still exploded every time it has flown). And Musk has some wild ass plan for catching the thing in midair with a hook on the launch pad.
My best guess for the end of the year is that Starship is flying some successful high altitude tests with non-explosive landings, but Super Heavy is at the “short hop” phase at best.
I also think 60% is optimistic, but do keep in mind that most of the Starship failures so far have been during part of the landing sequence, it is very possible that a test flight reaches orbit before either stage can be successfully recovered, in fact IIRC Musk has said that this will likely be the case. Still I'd personally put this prediction at more like 40%, rising to Scott's 60% if we extend the time-frame from Jan 1st to something more like April or May 1st 2022 (so a year from now rather than by the end of this year).
I don't think you've refuted Gbdub's point. Let's assume that Starship itself is done, finished, working perfectly, and completely ignore any landing requirements. To get to orbit, Starship still needs a working Super Heavy booster. The first Super Heavy production prototype was only assembled a few weeks ago: it'll take more prototypes, and plenty of development and integration work, before the full stack is working.
Scott's figure of 60% probability of Starship reaching orbit is, I think, about right for a window from now to the end of 2025. For the probability of reaching orbit by the end of 2021, I would put the probability at 2%. If you (or Scott) are amenable, I would be happy to bet on this, taking odds somewhere between these extremes for Starship in orbit by 31 Dec 2021. 10-to-1, say: you put up $X, I put up $10X, and the winner takes the pot. Interested?
I'm not the original commenter, but I'd be willing take that bet if you're still up for it. Say my $10 to your $100? Or higher, if 10 bucks isn't enough incentive.
I'll also offer a secondary bet of 1-to-1 odds that it makes it to orbit by end of 2023 (I think this is fair to you, if you give end of 2025 a 60% chance, but let me know).
Let me know if want to accept either, and we can work out exact terms.
Sure! Your $10 to my $100 sounds good to me: with those amounts, I'm happy to trust you for it, rather than worrying about an escrow account. Though I appreciate that you may feel differently about $100. We can discuss this.
I'll pass on the secondary bet: that one's close enough to my estimation that I could see myself being wrong (i.e. that a well-informed expert would take the other side of it).
Any idea how we can get in direct contact with one another without doxxing ourselves (e.g. posting our email addresses)?
You can email me at <my username> + n @ (google mail). Or through Reddit with the same username as here.
Emailed.
they got years to fogure it out.
1) Musk is a gormless poopyhead who can't do nothing orbital right.
2)Musk is a bombastic capable fellow who has launched a quarter of the satellites now in orbit.
Which is it? Really not an expert myself.
I mean, it’s kind of both. His publicly presented timelines are usually insanely optimistic, but so far SpaceX has (eventually) accomplished most of what it has set out to do.
Falcon 9 is a great rocket now, but it’s major milestones all occurred years later than what Musk originally claimed (still historically fast). And Starship is way more ambitious. Rockets are hard.
I mean, Starlink is basically on track from the timelines he was talking about around 2015. You can go look at public statements and such, and they're remarkably accurate.
Starlink is a much lower degree of difficulty.
Falcon 9 was announced in October 2005 for a first half 2007 launch, actually flying in 2010. Falcon Heavy was unveiled in 2011 with a planned first launch in 2013, eventually not launching until 2018. Red Dragon was scheduled to launch to Mars in 2018, as publicly announced as late as 2016, then canceled entirely. Falcon 1, Falcon 5, 2nd stage recovery, propulsive landing for Dragon, and dry recovery of payload fairings were all canceled after slick Elon presentations proved to be infeasible.
Again, SpaceX has a remarkable record of major achievements. But not every idea has been successful (not even ones that get as far as major hardware tests) and very few have happened on Elon’s optimistic timelines. And Starship is more ambitious than anything else they have attempted.
60% chance of orbit this year is very very optimistic. My median guess for a successful orbital test of the system is second half of 2023.
> Starlink is a much lower degree of difficulty.
While not disagreeing with the rest of your reply (I know nothing about SpaceX's history with deadlines, only Starlink's), is this really true? Proliforated LEO was still mostly a concept, not a reality back in 2015, and no one had ever mass-produced satellites in the way that Musk had proposed doing.
In hindsight, perhaps it looks less challenging, but I'd be interested to hear why, looking forwards in the early 2010s, Starlink appeared to be a much easier thing to execute than, e.g., Falcon 9.
Iridium did “proliferated LEO” in the 90s. Increasing the number of satellites by an order of magnitude or two is mostly a problem of affordable launch capacity (which Falcon 9 addressed), evolutionary developments in satellite technology, and the will to do it. Generally, doing lots of something you already know how to do is easier than doing something no one has done before - unknown unknowns are what get you.
It’s not that Starlink is necessarily easy, it’s that what Starship is trying to do is really really hard - the pure physics are much more challenging. Getting that many big rocket motors working in sync without blowing up is not at all trivial, let alone landing the thing on the same pad it took off from.
Propulsive landing for Dragon (after parachute, rather than sans parachute) wasn't human-rated by NASA, but as far as I know there wasn't any major unknown difficulty about it, as with catching fairings, which did turn out to be too difficult to be cost-effective for now after a number of attempts.
It wouldn't surprise me if propulsive landing for crew Dragon ended up significantly deprioritized partially due to the April 2019 failure. It was technically due to a different system, but I doubt there's much appetite to increase the volatility of crew capsules.
Then why also cancel Red Dragon, which was initially planned to be unmanned? I think some technical challenges e.g. Dragon getting heavier probably played a role as well.
Don't forget the original ITS timeline from 2016, with first unmanned Mars mission launching in 2022 and crewed in 2024. Not sure if Starship should inherit that one or if the whole system ought to just count as scrapped.
Posting predictions publicly to calibrate yourself makes sense. Not posting predictions about your and your friends' private lives publicly makes sense. The particular combination of those that you've chosen doesn't make sense.
Anyway good luck with Lorien. I'm very curious how that turns out, especially in this context since it's effectively the single biggest bet you've ever made. Really hope that it succeeds!
Maybe post the public predictions only? (Alternatively, I think it might be meaningful to precommit to the private predictions, then publish their SHA512 hashes or something along with their probabilities. Assuming SHA512 doesn't get cryptanalyzed into oblivion soon, this would allow one to keep private info private while still not leaving room to weasel out of past mistakes.)
I don't see how posting a hash is better than just posting "redacted" unless you plan to *eventually* make the thing public (allowing people to check that it matches the hash). Based on his grading of last year's predictions ( https://astralcodexten.substack.com/p/2020-predictions-calibration-results ) he doesn't appear to do that.
I have found that predicitng economics and finance stuff tends to be easy (GDP go up, stocks go up, rich ppl get richer, etc.) but outcomes of poltical events is way harder. This means you cna make easy $ by being bullish on stocks, most of the time, but prediction markets are more of a fool's errand.
Predictions that stocks will go up are right more often than predictions about political outcomes, but that doesn’t necessarily make them a better bet. It also matters what kind of odds you’re getting. With stocks, when they go down they tend to go down by a lot, so your implied odds are not 50/50. It’s like betting on a prediction market trading at 80c - if you bet “yes” you’re probably going to win, but that doesn’t mean it’s obviously a good trade.
I'm not sure I agree completely. It requires a deep knowledge of history, human psychology, and possibly the background of the individuals themselves, but - and especially in political games that rely a fair bit on populism - it is possible to be quite accurate.
Not to bray on my own behalf, but I predicted the mass protests we saw last year about a month in advance. I didn't know the political cause they would assume, but got the timing almost perfectly. We might counter that those were predictions of public sentiment and behaviour rather than pure politics, but that's exactly what opinion polls and the like are anyway.
15. Bitcoin above 100K: 40%
16. Ethereum above 5K: 50%
What's the reasoning behind these?
I subscribe to the idea that Bitcoin has a strong correlation (70%+) to loose US monetary policy. There is no indication that the Federal Reserve will be more accommodative than it is today, quite the contrary, most likely by the end of the year there will be signaling of tightening monetary policy in the near future, negatively impacting Crypto and other ultra high growth assets.
there is no correlation tbh. That is just a media narrative to explain what is just speculation. Bitcoin had its best year even in 2013 despite the fed signaling tightening.
Probably a fair bit of the reasoning is simply that BTC and ETH have already gone up a lot, so it’s not unreasonable to bet that the trend will continue.
It seems like he just expects the bubble to be a little likelier than not to fail to burts by year end. I suspect that he’s wrong about that, but only because my fiancé’s idiot brother tried to get her to open a crypto-trading account that isn’t available in his state so she could make straw purchases for him. That has “bubble near burst” written all over it.
"When my shoeshine boy was giving out stock tips, I knew it was time to get out." -Probably not J.P. Morgan or Joseph Kennedy
Scott, I’m half-trolling here, but do you have enough confidence in your BTC and ETH predictions that you’re willing to put money on them? I will take the other side of one or both, at whatever size you like. We can do the bet either for personal profit or for donation to a charity of the winner’s choice.
I've bought BTC and ETH, which seems like putting money on them enough.
How much of your net worth is in BTC and ETH, because your predictions represent a level of bullishness that is much more bullish than the market. You can back out these percentages from options markets, deribit is currently implying a 15-20% chance ETH will be over 5000 by years end, you said 50% I would say that if you don't have the majority of your net worth in crypto your predictions are unreasonably bullish
Why would having the majority of his net worth in crypto make a prediction that it would go up more reasonable?
I can see an issue with "if you're more bullish than the market and confident, you should Go Full Kelly", but that's a consistency/confidence issue, not a reasonableness issue. If I predicted that everyone would abandon Facebook tomorrow and shorted several million worth of their stock, I'd be consistent and confident but not reasonable.
Ok fine, consistent was more what I meant than reasonable.
Fair enough, then.
If you're going to put a significant amount of your net worth on the bet, it's also essential to factor in the likelihood that the value will tank. He doesn't make a prediction that the value will tank, so we don't have enough information to compute the reasonableness of doing that.
For example, maybe you think ETH will got o $5000 with 50% confidence, or will go to $1 with 30% confidence. Weird, but possible. Then, it's reasonable to put a small-to-moderate amount of money in, but probably a very bad idea to put the majority of your net worth in. Because if you put the majority of your net worth in, you're essentially going broke with 30% probability. You might think 30% of going broke vs 50% probability of getting obscenely rich is a great trade, but it's not, because money has diminishing returns. The the utility value of money is highly asymmetric.
All that to say, just based on the numbers we have, we don't know how good of an idea it would be for Scott to bet more on his BTC and ETH predictions.
You can back out a market implied pdf from options markets. Anytime your subjective beliefs disagree very strongly with this market implied distribution you can find a net that gives you incredibly good EV. I gave one example in my other post. At the time Scott posted this he could have bought an ETH call spread that would quintuple in value if eth reached 5000 by the end of the year. An investment with a 50% chance to quintuple in value is something you should be throwing lots of money at (37.5% of your bankroll by the kelly criterion).
Even if he doesn't want to mess with options and only bet through buying ETH he would have to have an incredibly unreasonable pdf on mind in order for it to both fit his predictions and not by an insanely good investment
"25. Greater than 66% of US population vaccinated against COVID: 50%"
Just to be clear, this means receiving tthe entire regimen of doses of a given vaccine and applies to the entire population, not just adults?
typo: I think you meant "ELON/TECH"
Getting a degree in Elonomics so that I can understand the Elonomy was the best decision I ever made. Most media attention goes to the macro elonomy, but I find Microelonomics more interesting.
What's your take on Elon is microelonomics versus Elon is a microcosm of microelonomics?
> "ELON/TECH"
> a degree in Elonomics
> the Elonomy
> the macro elonomy
> Microelonomics
The Elon Musk stans must be really extreme today.
(To be fair, I'm also a fan of Elon Musk - just not *that* extreme.)
Out of curiosity, why exactly are you working on separate essays for climate change and CO2?
Scott has talked before about CO2 having effects on human cognition at levels low enough to be anthropogenic (as low as 800 ppm). So if you believe anthro climate change risk is only moderate due to geoengineering being able to arrest any tipping point effects (atmospheric SO2), and take a Hansonian view of human welfare, and believe that high CO2 levels could lead to e.g. IQ declines on the level of 5+ pts globally then actually that effect is potentially far, far worse for humanity measured as an integral.
The CO2 essay sounds more like a follow-up to this essay, which is about the cognitive effects of elevated indoor CO2 levels, especially in poorly ventilated rooms: https://slatestarcodex.com/2018/08/23/carbon-dioxide-an-open-door-policy/
Very excited for the essays mentioned at the end; what can I say, I love the lurid topics like climate change. Also appraisers are paid to find complications; 50% is too low. Also linking to/outing DSL is probably not a great idea.
>Also linking to/outing DSL is probably not a great idea.
Disagree, what would be the problem specifically? Despite the name I don't believe the forum is attempting to be a secret or anything.
I guess I'd say it has a tendency to indulge/skew toward the kind of direction Scott has tried to avoid his community going in, historically. cf More Right, The Motte, etc.
Your comment made me go search for the link (I'd previously assumed they were all just links to pages in his own blog) and follow it. An amusing mini-Streisand effect.
Agreed. Scott should avoid signal boosting DSL unless its absolutely necessary
> 30. Some new variant where no existing vaccine is more than 50% effective: 40%
Recommend you clarify this. How certain does the evidence need to be? How prevalent the variant? Like, do you get this right if there's a variant that some case study finds 10 cases of and 6 of those people die? I think you technically would, but I think it would be pretty irrelevant.
Another way of thinking about it: I'm reasonably certain that this will be true for "some variant", so the question is really about whether it matters enough for anyone to notice. If you're really looking to predict something about the prevalence of a variant that is highly resistant to the vaccine, you should re-word/clarify.
> 8. Tokyo Olympics happen on schedule: 70%
I hate to break it to you, but...
I guess he means the schedule according to current plans?
I'm sure he does, but I think we've seen that it pays to be crystal clear in these threads. :-)
Scott, I will take the other side of your Dow Jones predictions in unlimited size. The delta on the DIA 350 12/31/2021 call (roughly the probability of the DJIA ending above 35k) is 0.417 - you're way way off market.
Delta isn't the probability of an option ending up in the money. If you want to see what the market is implying you need to look at the price of a very tight call spread as this approximates a binary option
OK, so if I take the mids of the 350/355 DIA 12/31/2021 call spread, I see it a 2.25 debit which would be.... very close to the delta calculation no?
Yeah it's pretty close
What do you consider to be BLM protests? Do any anti-police protests count, do they have to be endorsed by any particular group, or are you just counting any protest that is generally aligned with the goals of BLM?
"Court packing is clearly going to happen"
Who defines clearly? "Scott Alexander thinks court packing is going to happen" or "The popular consensus (as assessed by Scott Alexander) thinks court packing is going to happen"?
I assume "everyone (not just most people, but like everyone that is paying attention and isnt stupid) agrees it is going to happen"
I think this mostly comes down to whether they change the law that limits the size of the court. If they change the law to allow, say, 13 justices, then it's basically a sure thing that they're going to go ahead and appoint those extra justices.
Needs Manchin's approval, probably won't get it. Also needs filibuster abolition, right, since it's not budgetary? And Manchin has said no on that too.
Are you counting all coronavirus variants for your covid predictions? If not, I feel like you should be more precise.
Would be cool to put out these predictions first for a bit before showing your probabilities. I'd be interested in assigning my own without being biased by yours.
I don't see any major BLM riots. Police chiefs are making sure their officers aren't doing things that look horrible on video, and we now have a presidential administration that knows how to reduce tension in the black community, rather than inflame it. I see small scale protests whenever a black person gets killed by police, but nothing like what happened last year.
What did the Trump administration do in the immediate aftermath of George Floyd's killing that inflamed tension in the black community?
Many things, but off the top of my head he tweeted things like "when the looting starts, the shooting starts" and sent in Federal law enforcement against the wishes of local officials.
Had there already been major rioting by the time Trump tweeted that (dangerous and disgusting comment)? As a Trump-hater, I'd *like* to be able to assign him a decent share of riot-blame, but if they were already serious by the time he said that, seems wrong to blame him or predict low-intensity this time.
Nonetheless, my gut instinct is that Scott is a bit high on that one.
> we now have a presidential administration that knows how to reduce tension in the black community, rather than inflame it
More to the point, the Powers That Be don't want there to be riots under a Biden administration, so there won't be. Riots don't just happen.
I'm sympathetic to this line of reasoning, but Portland has been rioting for weeks now, Biden notwithstanding.
I mean, it's Portland – they're probably always going to be exceptionally likely to continue 'The Resistance' under any plausible Democratic Party presidential administration anyways.
This is true, but it speaks against the 'powers that be' hypothesis.
True! But then my personal prior is that Portland would likely be an exception to any 'powers that be' (in their own minds at least).
I'm kind of impressed you view the odds of a major problem in the China/Taiwan issue at only 5%. Most watchers I've read are nervous about that, with the feeling that China is feeling its oats and the US has been somewhat prostrated by COVID and cultural conflicts, so...the time might be right. I'm a little curious whether you read deeply on the subject and decided that's all worrywartism, or whether this is more along the line of "well nothing's happened in the past 50 years of occasional concern so..."
I mean, IIRC Scott lives in the Bay Area (certainly on Moscow's list, and almost certainly on Beijing's and Pyongyang's), which kind of implies that prediction. If someone's bet their life on ~WWIII, it's kind of obvious that they don't think WWIII is very likely.
(You could get cynical here and say that given he's already betting his head there's no marginal risk in betting his internet reputation as well, but I don't think Scott's that mercenary.)
A fight between China and Taiwan won't necessarily involve the U.S., and even if the U.S. does get involved it won't necessarily escalate to nuking cities.
I mean, not *necessarily*, but it seems extremely likely in the conditional. I think after Hong Kong anybody who let Taiwan go without a fight would feel like Chamberlain, and while the USN is very mighty it's probably not up to fighting within range of PRC airfields. The US's ability to dictate terms, AIUI, consists pretty much solely of their ability to win a nuclear war via larger arsenal + allied close-range launch sites.
But what are the odds of an amphibious assault within the next 8 months? (really 6 due to unfavourable winter weather) They'd have to make the decision within the next few months if they want to be ready in time for the October window.
Over a year passed between the Trident Conference when the allies agreed on a cross-channel invasion and D-Day.
Well, not high, but I suppose I thought "major" event could fall short of a full-scale invasion and still qualify. To my mind "major" is probably anything that genuinely involves the USN. I'm pretty sure we (the US) could and would overlook any number of Taiwanese casualties that *didn't* imply an actual takeover, but if for example a US ship got hit that would be difficult to rationalize away, much as the current (or previous, honestly) Administration will want to do so.
I don't think a major event short of full-scale invasion is all that likely, since it's not in the interest of the PRC.
The 7th Fleet and ROCA are both kind of badly run. A major act of aggression from the mainland is likely to create the political will in Taipei and Washington to go in and fix these problems. It'd also make it easier for Taiwan to buy equipment like diesel-electric subs and modern air defense systems from other countries.
Taiwan is buying Patriot, wants to buy THAAD but that would be almost a red-line event for China, and is building their own modern air defense systems as well. Taiwan cannot buy diesel-electric subs from anyone because literally no nation capable of building diesel-electric subs is willing to sell arms to Taiwan.
What I mean is that if the PLA does something that is widely considered to be a dick move e.g. taking Kinmen, it'll shift the geopolitical landscape. Many countries will become more willing to sell arms to Taiwan and the big bad PRC will have less standing to complain about the ROC buying more modern systems.
In terms of realpolitik, a sudden aggressive move by China would anger a lot of countries at once and It'll be harder to push them all around with trade sanctions etc. than it would be to punish an individual country for getting too cozy with the ROC.
I've seen the mentions of PRC figures saying "we'll go in if Taiwan gets THAAD".
Can someone explain to me why they've got a problem with Taiwan having THAAD? They don't exactly need ballistic missiles to attack Taiwan - it's within range of bombers and of cruise missiles - and as it's not boost-phase it doesn't degrade their deterrent against the USA.
Am I missing something?
Their claim is that the THAAD radar is secretly meant to perform boost-phase tracking and battle management in a way that does degrade their deterrent against the USA. I'm skeptical that they really believe this, but they do say it when anyone tries to put a THAAD site anywhere near China.
I think it's much more likely they believe that yes, they do need ballistic missiles to attack Taiwan (or South Korea or Japan or anyone else). Or defend against an attack coming from those places. Yes, those places are within reach of China's bombers and cruise missiles, but those places are protected by air defenses which can shoot down China's bombers and cruise missiles. And China doesn't have Wild Weasels. Their edge comes from having lots of effective precision-guided ballistic missiles and their targets *not* having effective missile defenses.
You seem to assume that they haven't already made the decision. Is there some reason we would know if they had? The US military intelligence apparatus would probably at least suspect, but my understanding is that they don't publicise that sort of thing right away.
Some of the rhetoric at the Alaska summit (specifically, "Taiwan is a domestic issue", "we will have to take firm actions" and "if we confront each other, it will only damage the US; China will pull through") could be taken to mean "we are going to invade Taiwan; stay out of it or you'll lose".
Wait you think supreme Court packing is just as likely as Newsom being recalled?!?! Surely that's a typo.
Or did you make that prediction genuinely on the 1st? it's a bit unclear from the post when the predictions were made.
Which of these are you suggesting is more likely? Both seem plausible but unlikely.
1) the Democrats have the ability to court-pack if they want, plausibly could run into a situation where it would benefit them (i.e. they're not missing a house or getting vetoed, but have an unfriendly SC and don't have the numbers to go the impeachment route), and publically threatened it after Amy Barrett was nominated. However, it seems relatively unlikely that they'd actually pull the trigger.
2) The petition to recall Newsom succeeded in getting the proposition on the ballot, but he's recalled if and only if the proposition passes. Current polling is at 60:40 or so in his favour from my digging, which is a large but not impossible gap to make up.
"[redacted] wins the book review contest: 60%" made me wonder. It seems unlikely that this is referring to one specific review, since it would have to stand out massively in a contest of some 20+ reviews (assumption based on "I’ll be posting about two of these a week for several months"). It could be a categorical statement like "A fiction review wins the book review contest", but that seems too easy. My prediction is that this is commentary on a fundamental issue with contests like these, for example how people are statistically more likely to choose whichever option was presented last as their favourite.
I agree that 60% seems high, but I would guess that is indeed for a specific review. Bear in mind that a) Only Scott has read all the reviews b) Scott probably knows already which review he's going to vote for, and c) Scott's choice will have (probably) the same weighting as everyone elses votes put together. Or something.
Or put it another way. If Scott granted only a third of the voting power to himself (on his own blog, running his own book review competition, and giving away lots of his own money) it would STILL be quite remarkable if the commentariat managed to disagree enough (and be unanimous enough) to outvote him. Probably also true if he only has a quarter of the votes.
In fact, 60% seems quite unconfidant...
Maybe the voting format is going to be different from my expectations.
Back in Open Thread 166 Scott said:
>Somebody has to choose the winners, and it's either going to be me or the readership. I would rather people write things my readers like than things that I like, especially since I'm easier to manipulate (you all know what kinds of things press my various buttons).
You could interpret that various ways, but I think that means he's not going to reserve significant voting power to himself. He *did* already have a rather significant hand in choosing the winner (via picking the finalists), after all.
Fair point. I was vaguely remenbering some much earlier comment about how Scott and the commentariat would both have an input into deciding the winner(s). The comment you quote makes Scott's prediction much more understandable.
Aside from what Anteros said, it's also sometimes the case that things can be heavily favored with only a small difference in "quality" --
Think Biden being favored over Trump 80-20 by 538 in the 2020 Election, even though he was only a few percent better in the polls, or a sports team being heavily favored in a playoff series ever though their record is only 10% better. (Neither of these are exactly the same as the book review contest, of course)
Perhaps this review only stands out a little bit, but if that slight difference in quality is clear to a good percentage of voters, that could be enough to make it strongly favored.
I have added these to a foretold community (https://www.foretold.io/c/6eebf79b-4b6f-487b-a6a5-748d82524637), and predicted on the non-redacted ones. It might be a good place for the commentariat to make their own predictions. Making predictions requires an account, but viewing them doesn't.
I've also added the predictions from that community to metaforecast (https://metaforecast.org/?query=%20&forecastingPlatforms=AstralCodexTen&numDisplay=200), which provides a somewhat nicer interface (i.e., you can search for particular questions)
Because the number of questions is relatively large, (75 non-redacted ones), it would be relatively simple to have a tournament, with a relatively simple scoring rule/winning criterion.
"Significant capital gains tax hike (above 30% for highest bracket): 20%"
I'm assuming you mean national rates. California has already exceeded 30%.
"India's official case count is higher than US: 50%"
This seems quite low. India already has half as many cases as the US, and their healthcare system is collapsing.
Hard to count when the healthcare system is collapsing.
Looking forward to the reviews of Nixonland and How Asia Works. Not to snub The Scout Mindset, it was a valuable book that I'm more likely to recommend to friends than the previous two, just not as excited to see your take on it.
You can meditate with a traditional mindfulness technique for 3 minutes a day effectively and have a far greater accumulative impact than doing 10 vippasina retreats or squirming in fits of boredom alone with an app never knowing if you’re making progress. More importantly - Direct transmission from a realized lineage holder/teacher > apps or secular modern neuroscience based meditation and its rigid container. Exponentially more profound realizations of time/form/phenomena/mind/afflictive emotions vs simplistic stabilization is avail w a lineage teacher, group, setting. The focus through breath (Annapurna) or even impermanence insight meditation is scrunched into discursive mind without deeper containers.
Or.. one can just sit and do Metta/emptiness meditation properly with open heart. Even for a minute. Actual liberation is experienced rather than parlour games of dilating focus through a misapprehension of mind and causal interdependence. Just saying! :)
Bleh – the main thing that people that are a "lineage holder/teacher" ('guru') seem to be good at is being (or pretending to be) an 'impressive lineage holder'.
The people that actually seem to be impressive or successful, or even 'just' _enlightened_ – in my experience anyways – seem to strongly discount the potential impact of meditation.
(I also suspect that almost everyone regularly engages in something that's functionally equivalent to meditation.)
Most of the supposed benefits – which are _traditionally_ all self-reported – seem more likely to be the placebo effect (i.e. wishful thinking).
And any honest (and accurate) accounting of the overall effects of meditation practice needs to grapple with The Dark Side of Meditation. I've never seen anyone try to reasonably offset that and even roughly determine whether meditation is good or bad on net.
Bleh – the main thing that people that are a "lineage holder/teacher" ('guru') seem to be good at is being (or pretending to be) an 'impressive lineage holder'.
---You've not shopped around sufficiently. Or your'e parroting an anti teacher trope favoured by the western degradadation of Buddhism and contemplative work in this era. Yes, egos abound with teacher types. But... egos abound period (and to all types of degrees and qualities.)---
The people that actually seem to be impressive or successful, or even 'just' _enlightened_ – in my experience anyways – seem to strongly discount the potential impact of meditation.
--That's curious. Unless someone is realized at birth, or through traumatic intervention or an amazing confluence of events resulting in the ability to directly realize mind/interdependence and emptiness etc etc... I think you're confusing enlightened with..morally upstanding or wise. They are not separate but neither are they always (even often) the same.--
(I also suspect that almost everyone regularly engages in something that's functionally equivalent to meditation.)
Most of the supposed benefits – which are _traditionally_ all self-reported – seem more likely to be the placebo effect (i.e. wishful thinking).
Yes, we should jettison 5k years of vedic and western contemplation and texhniques, profound cannonized teachings based off of them and the direct expression of millions of humans today. All just delusional yearning into the void and conformation biasing some deeper confusion? No.----
And any honest (and accurate) accounting of the overall effects of meditation practice needs to grapple with The Dark Side of Meditation. I've never seen anyone try to reasonably offset that and even roughly determine whether meditation is good or bad on net.
-- You/one will experience the shadow on the way to the light. We are not after happiness as it is understood through the sensual lens. That is the obstacle. We are after liberation and it includes a very dangerous path of recognizing how corrupted our perception and reasoning is. That is what one discovers. Yes, it is traumatic. And beyond that resides a joy. A true joy. I ain't trying to sell you anything. :)
> You've not shopped around sufficiently. Or your'e parroting an anti teacher trope favoured by the western degradadation of Buddhism and contemplative work in this era.
I would imagine that you're so sure of this because you're enlightened, huh?
> I think you're confusing enlightened with..morally upstanding or wise. They are not separate but neither are they always (even often) the same.
What's the point of enlightenment if it doesn't also mean (at least) _more_ moral or wise? The Western movement with the same name was, and I think 'properly', very much concerned with both morality and wisdom (even if, in practice, it fell short of either, or both).
> Yes, we should jettison 5k years of vedic and western contemplation and texhniques, profound cannonized teachings based off of them and the direct expression of millions of humans today. All just delusional yearning into the void and conformation biasing some deeper confusion? No.
I didn't write or even imply your summary.
I do think a _lot_ of it is something like "delusional yearning into the void and conformation biasing some deeper confusion".
But it's a big topic/subject/area-of-experience! There's, as you point out, _thousands_ of years of activity around this. There definitely _is_ real, significant, insight, wisdom, and even ethical and moral efficacy, tho sadly it's, like all traditions, 'protected' as sacred without being understood.
> You/one will experience the shadow on the way to the light. We are not after happiness as it is understood through the sensual lens. That is the obstacle. We are after liberation and it includes a very dangerous path of recognizing how corrupted our perception and reasoning is. That is what one discovers. Yes, it is traumatic.
I mean sure – I appreciate just the _idea_ of things like 'charnel ground practice'.
And I don't think "happiness" is a proper ultimate aim (or 'terminal value') either.
But I also don't think 'happiness' is an "obstacle", let alone _the_ obstacle, and I think aiming for "liberation" is, if anything, WAY more stupid (and evil) than aiming for 'happiness'. 'Liberation' in a metaphysical/mystical/spiritual sense can only mean one thing – death. A truly enlightened nihilist would promptly die of dehydration (not that the actual death would arrive promptly).
Some amount of liberation is perfectly fine!
I don't think existence is something from which beings need to be generally 'liberated'.
And you write "how corrupted our perception and reasoning is" whereas I think that's silly and obviously 'charnel ground practice'. I think the boring/mundane truth is just that ANY perception or reasoning, by any kind of 'being', must be boringly/mundanely 'implemented' _some_ way, and _one way of looking at_ that brute (and Sad) fact is that any perception or reasoning is necessarily "corrupted". I think 'limited' is a more enlightened perspective.
I don't think The Truth is traumatic either – it sure is for some people tho, I'll admit. But maybe that's just because we're all generally confused about what it is, or how best to reach it (or whatever peak(s) we've already discovered).
> And beyond that resides a joy. A true joy. I ain't trying to sell you anything. :)
"A true joy" huh? <sarcasm>Yeah, I have NO reason to suspect you of selling me a wireheadseat.</sarcasm>
> Major flare-up (significantly worse than anything in past 50 years) in China/Taiwan conflict: 5%
Just a reminder that the past 50 years include:
- the PRC and ROC militaries exchanging symbolic artillery fire with propaganda leaflets on alternate days during the 1970s
- the 1987 Lieyu massacre, during which 19 Vietnamese boat people were killed by the ROC military after stranding on a militarized island near Kinmen. (Kinmen and surrounding islands have since been put under civil administration starting in the 1990s.)
- the Third Taiwan Strait Crisis (1995-1996) when the PRC military fired ballistic missiles into the waters near ROC-controlled islands, including Taiwan.
- PRC military aircraft crossing the center line of the Taiwan strait in 2021, but at an angle so they were not flying directly towards Taiwan.
My guess is that when establishing the baseline of "significantly worse than anything in past 50 years," you were thinking about some, but not all of these events.
My alternative predictions:
- PRC military aircraft cross the Taiwan strait at least two thirds of the way, on a course headed straight towards Taiwan: 5%
- PRC military conducts live-fire exercises in waters near ROC-controlled islands in a repeat of the Third Taiwan Strait Crisis: 2%
- PRC artillery hits land controlled by the ROC: 1%
- civilians are killed by troops fighting in the Taiwan Strait: 0.5%
- PRC gains control over Taipei: 0.1%
- PRC, ROC and US militaries continue to make plans involving a PRC invasion of Taiwan as a possible scenario: 99%
I know very little about the specific details of the China/Taiwan situation besides the knowledge that comes from reading mainstream news articles, plus the book On China. But I do think that when making probabilistic forecasts of any kind, it’s very difficult to justify estimates less than a few percent. I think the world is just generally far more unpredictable than people think. So on that basis, I would argue that your probabilities should not tail off as aggressively - maybe they should ramp from 5% down to 1% or 0.5% rather than 0.1% (with intermediate scenarios adjusted proportionally).
0.1% is easily reached by a combination of three factors with 10% chance each. It's probably not very helpful to include events that unlikely when making only a few predictions, though.
For the PRC to annex Taiwan, it'd have to invade and the USA would have to decide to allow it (the PLA is not sufficiently powerful and decentralised to eat several hundred nukes and then mount a contested amphibious invasion against 1.7 million Taiwanese reserves - outright invasion of China is probably beyond immediate US capabilities, but stopping a boots-on-the-ground invasion isn't). Predicting 2% invasion/1% PRC takes Taipei is implicitly predicting 50% conditional of "USA allows Taiwan to be invaded", which seems way too high (it's not like it's a domestically contentious issue; the one major thing Tumblrites and Trumpists agree on is "Beijing bad", and allowing a US ally to fall uncontested would be permanent political and legacy suicide).
I agree. I think an estimate around 1% is more reasonable than either 5% or 0.1%.
Integrated over all "plausible conflicts" (which are maybe in the order of tens) it would mean that there is one conflict every few years that breaks out without it being majorly foreshadowed, which sounds about right. Although I am a little skeptical of that line of reasoning, as state-vs-state conflicts nowadays have higher hurdles, and if you only count state-vs-state scenarios you have much different base numbers.
What's your prediction of "PRC invades Taiwan"?
made the mistake of thinking Scott was prediction an approval rating for Biden of between 50% and 80%. Fortunately, I caught it before I posted to say Biden wasn't going to get an 80% approval rating, but I do think the prediction was easy to misread. It needs more space between the two percentages or something. Or prediction percentages should be a different color.
Also, I don't see the point of giving each redacted prediction its own line. If you want to mention them, how about just saying how many of them there are in each category?
Why wouldn't the US approve the AstraZeneca vaccine? (I'm not in the USA and I don't see how this is reported there).
That stood out to me too.
Why wouldn't they? Why haven't they?
The FDA doesn't like approving things. Even now when I got the vaccine I had to read an FDA statement saying that they were not approved, just allowed because of an emergency. So I think betting that the US won't approve a vaccine is generally a good idea.
The United States doesn't need the AstraZenica vaccine for its own purposes, and it takes substantial political capital to make the FDA approve anything in less than years. So the question is, how much political capital is the Biden administration willing to spend to be able to donate vaccine to India? Not zero, certainly, but also not certainly enough to get it done.
minor complaint regarding the conflict predictions. "Significantly worse" leaves a lot of room open to interpretation.
I think Scott does sometimes throw out predictions if the result is hard to call one way or another. Hopefully this will be one of those where there will be basically no conflict and it's easy to judge...
> 21. Google widely allows remote work, no questions asked: 20%
You probably mean specifically fully time employees who are knowledge workers at Google, correct? In this case it might be useful to add more contextual qualifiers to your predictions at the risk of making the list a bit wordier.
Am I insane or was there not already a review of Nixonland published here? I just looked in the archives, though, and this post was the only one that came up.
Yep, that was it. Thank you. Not sure why it didn't come up in the search I did earlier, though.
63. reminds me that the US is still a dystopia that accepts paper walls, flimsy houses in general and incessant TV consumption as completely normal.
Thank god for the 50+cm of solid stone all around me... No AC necessary, either!
I've recently learned about Heat Pumps and their incredible efficiency at heating and cooling no matter the surrounding temperature. Although I couldn't find numbers for how efficient they are considering the transport waste of electricity... Promising nonetheless!
Your comment reminded me of G. K. Chesterton's shock when he visited the American countryside and found that all our houses were made of wood! Before I read that account I never thought there was anything particularly remarkable about a wood frame house.
"The sharpest pleasure of a traveller is in finding the things which he did not expect, but which he might have expected to expect. I mean the things that are at once so strange and so obvious that they must have been noticed, yet somehow they have not been noted...I had no sooner passed out into the suburbs of New York on the way to Boston than I began to see something else quite contrary and far more curious. I saw forests upon forests of small houses stretching away to the horizon as literal forests do; villages and towns and cities. And they were, in another sense, literally like forests. They were all made of wood. It was almost as fantastic to an English eye as if they had been all made of cardboard. I had long outlived the silly old joke that referred to Americans as if they all lived in the backwoods. But, in a sense, if they do not live in the woods, they are not yet out of the wood.
…
These things seem somehow to escape the irony of time by not even challenging it; they are too temporary even to be merely temporal. These people are not building tombs; they are not, as in the fine image of Mrs. Meynell's poem, merely building ruins. It is not easy to imagine the ruins of a doll's house; and that is why a doll's house is an everlasting habitation. How far it promises a political permanence is a matter for further discussion; I am only describing the mood of discovery; in which all these cottages built of lath, like the palaces of a pantomime, really seemed coloured like the clouds of morning; which are both fugitive and eternal.
…
I was always told that Americans were harsh, hustling, rather rude and perhaps vulgar; but they were very practical and the future belonged to them. I confess I felt a fine shade of difference; I liked the Americans; I thought they were sympathetic, imaginative, and full of fine enthusiasms; the one thing I could not always feel clear about was their future. I believe they were happier in their frame-houses than most people in most houses; having democracy, good education, and a hobby of work; the one doubt that did float across me was something like, 'Will all this be here at all in two hundred years?' That was the first impression produced by the wooden houses that seemed like the waggons of gipsies; it is a serious impression, but there is an answer to it. It is an answer that opens on the traveller more and more as he goes westward, and finds the little towns dotted about the vast central prairies. And the answer is agriculture. Wooden houses may or may not last; but farms will last; and farming will always last.
What is Prospera? I've searched and searched and just can't figure out what it might be.
A cool city - go back half a dozen articles on this blog and Scott has an article on it.
Got it! Thanks!
https://astralcodexten.substack.com/p/prospectus-on-prospera
Scott, who do you follow on Twitter?
6. Significant capital gains tax hike (above 30% for highest bracket): 20%
Seems too low, due to democratic control of everything at the federal level, plus corporate lobbyists not really caring one way or the other about individual tax rates, plus an unprecedented deficit and debt/gdp ratio and rising interest rates on US treasury bonds for the last ~8 months since the bitcoin bull market began (negative real interest rate bonds are a sucker bet that will lose market share to crypto)
7. Trump is allowed back on Twitter: 20%
Seems waaaay too high. The prior is that twitter almost never unbans anyone (especially not months later -- I haven't seen a single instance of that -- errors are corrected within a couple weeks or never). Plus the social dynamics inside twitter would make it even harder to unban him. It's full of woke employees, so championing the cause of free expression as applied to that particular case would be career suicide there. The democrat controlled congress certainly isn't going to pressure twitter to unban trump either. I'd say 1%.
The likelihood of #29 is probably higher than 50% on a standalone basis.
But 40% for #30 almost requires a higher % for #29: If a new variant evolves for which current vaccines are less than 50% effective, it will surely be very widespread relative to existing variants and therefore likely responsible for >25% of cases.
The two exceptions are if there are multiple variants like this (therefore preventing either from claiming a large share on its own) or if such a variant appears very late in the year, thereby fulfilling #30 but not having enough time to reach a large share.
>Ethereum above 5K: 50%
the DEC31 3840-5000 call spread is offered at 240 on deribit right now, that spread is worth strictly more than a binary option that pays out 1160 if this prediction is true. That is to say you are giving something a 50% probability that you can bet on at 4:1 odds for large size. That kind of disagreement with the market is bold, not necessarily wrong but bold. If you stand by this prediction in light of how stongly the market agrees you should probably try to get set up to trade crypto options, its a pain as a US citizen but not impossible.
*market disagrees
> Fewer than 10K daily average official COVID cases in US in December 2021: 30%
Ouch, that's within a factor of 2 of where we are right now, and 70% chance we'll be worse than that. So you're really thinking this isn't going away anytime soon. I was hoping we'd see it dwindle to where we could all ignore it, especially for the sake of the kids too young to get vaccinated (and who are unlikely to die, but might get damaged in poorly understood ways).
I'd be interested to hear more about how you think the next year is going to play out. (Especially combined with your pessimism about vaccine escape).
"Court packing is clearly going to happen (new justices don't have to be appointed by end of year): 5%"
I assume that this simply means "Court packing is ... going to happen: 5%." Otherwise, I'm confused about what the 'clearly' means here. Is there then a separate prediction for it *not* clearly happening?
I think the point here is that court packing hasn't yet happened by 2021 end, but decisions have been made, as of 2021 end, to pack the court after 2021 end.
Right; even if the Democrats manage to make "court packing" happen, there's likely going to be a delay while Democratic progressives try to pack it with maximally-liberal justices and Joe Manchin says "that's not what I signed on for; try again". Or maybe Biden is a gracious winner and only nominates respected center-left judges for quick confirmation, but it's reasonable for Scott to give himself some wiggle room on that.
Scott, what is the "current exercise routine" you mention in number 67?
> 105. I finish and post the CO2 essay I'm working on
I would like to humbly suggest that you might find some useful information about CO2 on the blog Wonky Thoughts by Doug Robbins. He writes infrequently, but the posts contain a lot of data.
- https://dougrobbins.blogspot.com/2019/12/understanding-source-of-rising.html