I'm going to try to think of all the ways putin can leave office in the next 11 months and quantify each of them using mostly the outside view. * anyone invades russia and overthrows putin: ~0%, because nukes. This has never happened to a nuclear country. * putin dies of any natural cause: 2%, because of actuarial tables: https://www.ssa.gov/oact/STATS/table4c6.html * putin is assassinated: 1%. If we count Stalin as half an assassination because he was probably poisoned with an anticoagulant but this has never been totally proven, then we have a ...
It took a lot more than fifty years for the population of the British colonies in north America to scale from 1 to 1M, despite the voyage being much quicker/cheaper, huge fertility rates, the economic draw of commodities produced in the new world, and the push factor of religious minorities wanting to escape persecution in Europe. Musk has stated a target of $10/Kg for earth->LEO launches using starship. Let's say the unmanned Mars->Earth cost is about 10x that (in terms of energy it's much cheaper to get off the surface of mars, but then the Hohmann t...
I think the difference between Russia's total success in Ukraine and total failure in Ukraine is probably only a few percent on this market. Either way, a coup is very unlikely. It's been 31 years since the last attempt, 56 years since the last success, and over 100 years since the second to last success. Throughout almost this entire period Russian leaders were doing horrible things that would have warranted a coup. The USSR lasted through ten years of an Afghan quagmire without a coup. The present moment doesn't seem special. I think Putin is mor...

Unfriendly superintelligence: ?% All other causes combined: <1%

60 million years since the MRCA of primates, versus at least a billion years till the increase in the sun's luminosity makes Earth uninhabitable. But we wouldn't be having this conversation if we hadn't involved intelligence, so 50% per 60 million years is probably a very upward biased outside view. I think the great filter is likely behind us (maybe abiogenesis, maybe the transition from primitive bacteria to eukaryotes which took 2 billion years, or from eukaryotes to multicellular life which took another 1.5 billion years) My wild guess is a const...
https://www.reddit.com/r/conspiracy/comments/r45a5n/here_is_the_evidence_that_reddit_user_maxwellhill/ Prior: 0.6/430000000. probability that Ghislaine has a reddit account / total reddit users. Evidence: Maxwell surname: *40000 Family estate in a place with "hill" in the name: *50 Birthday after Dec 21: *365/9 High-IQ power user: *50 Made no posts for 3 days around the time of Ghislaine's mother's death and around the time of the Kleiner Perkins party: *10 "Their coverage noted that Maxwellhill “uses a mix of British and US spelling as wel...

I actually lost points updating to 98% when the consensus was 94%, only because I didn't update to 99% again an hour later. The point system needs to be redesigned somehow.

Outside-view base rate: no change in 152 years. Assume E(continuation years) = 152. 29/152 = 19%. Changes before that were mostly due to creating new circuits, before circuit courts existed. If it happens now it will be for partisan gain. A couple were attempts to push partisan ends: 1. Lameduck federalists in 1801 (repealed after 1 year). 2. radical republicans gerrymandering the judiciary in 1866 to screw former confederate states who still lacked representation to oppose the change. Alternate base rate: 2 times in the 233 years since the c...
Ordinary flipping the incumbency advantage around would put someone at a huge disadvantage for non-consecutive re-election. cdo256 below calculated a historical base rate of 11%. But I think covid was a big factor weighing against Trump in 2020. Without covid, the booming economy and lack of emergency measures to expand vote-by-mail would have made Trump's re-election 80% likely. But not much about the pandemic changed because of the new administration, and the pandemic will be less salient in 2024. What will be salient in 2024 is probably high i...

The ACS report is a dead link.

"will any of these elements ever, even temporarily, increase 10x in price" is a much broader criterion than the headline implies. If mining operations are concentrated, a war could temporarily 10x prices. Or a surge in demand due to the discovery of new uses could 10x prices until mining catches up (like what already happened to palladium)

I think the headline would be more accurate if the requirement was that the price stayed at max($500/oz, 10x) for at least a decade.

How I calculated 6%: 10 out of 535 senators+representatives are openly LGBT. 4 out of 500 Fortune500 CEOs are openly LGBT. Let's split the difference and call it a 1.33% chance of electing an LGBT president in a wide-open election year right now (and much less than that when an incumbent is up for re-election). Gay acceptance is already so widespread that there's no incentive to be closeted (quite the reverse). They're underrepresented at the top due to personality differences (gays and lesbians are on average are less masculine than hetero men) and due...

Presidents of china ages upon ceasing to be president: Hu Jintao 71 Jiang Zemin 76 Yang Shangkun 85 Li Xiannan 78 Soong Ching-ling 88 Mao Zedong 83

Xi Jinping will be 77 when this resolves, which is younger than 2/3 of the prior ones. So the outside view gives us 66% which agrees with the community prediction.

The price on an individual exchange can get way below the general market price for an instant when somebody gets liquidated. But I think the spirit of the question is the general market price of bitcoin. To disambiguate it should declare the settlement source is an aggregate of several exchanges (e.g. coinmarketcap.com if that still exists)

Updating to 16% because: * sanctions on Russia really can get a *lot* worse * Russia is diverting troops to Kazakhstan * Putin has a much weaker pretext now than in 2014. Crimea used to be part of Russia until 1954; Crimean ethnic-Russians outnumber Crimean ethnic-Ukrainians four to one, and Crimea voted overwhelmingly to join Russia. (International reaction to the Crimean independence election has disappointingly been to basically deny that the Crimeans had any right to an election in the first place; and to disparage the integrity of the election ...

8 days have passed with no further corroboration of the coup claim. It's probably bogus. The blood cancer might kill him this year though. I'm updating from 95% to 85% because of the blood cancer and his appearance of ill health.

— edited by JonathanRay

@evanbd Lost points in absolute terms. I predicted lower than the community consensus from Dec 22 to Feb 11, and then predicted higher than the community consensus continuously after Feb 11. I think I understand why now. It's doing a time-weighted-average of the point values of all my predictions?

@(gbear605) He had a heart attack in 2012. Prior heart attack reduces life expectancy by more than half in his age bracket. He is still very fat for a Chinese person. I took the per-year death numbers for the social security actuarial tables, and pasted them into a spreadsheet to perform a calculation. Unmodified, these numbers imply he has a 22.1% chance of dying in the next 8 years. If we assume his per year death chance is doubled by the prior mild heart attack, he has a ~40% chance of dying in the next 8 years. At his age there's also a lar...
I just calculated that the same transistor count as a pentium 2 (7.5 million) could now fit on 0.043mm^2 with TSMC's 5nm process. But factories that make TSMC 5nm have a 2*10^13 times bigger footprint (950,000m^2): https://www.anandtech.com/show/12377/tsmc-starts-to-build-fab-18-5nm-in-early-2020 lg(2*10^13) = 44 79/44 = 1.8 To get any decent self-replicating nanobots at all by 2100, we'd need to halve the size of the factory every 1.8 years. But there doesn't seem to be any trend in that direction. To make transistors smaller, the factories get b...

If necessary, Elon can rescue it with his personal tens of billions of cash that he got from selling Tesla stock.

updating all downward on time decay and the Jones No tweet