@Jgalt On the upside, it's definitely a related question.

So far as I can tell, no one is expecting stormtroopers with laser guns. The question is about the creation of a new arm of the US military with the title of "Space Force" or something similar. Maybe that'll happen, maybe not, but the creation of a new branch of the military is not likely to be a purelty incremental affair; at some point there will have to be some sort of official declaration that there is such a thing, and that's what the question is about. The question isn't asking "when will there first be any funding for any sort of military use of ...

@alexanderpolta Honestly, if we had a direct report from one of those vanishingly rare individuals, I think (p=0.75) you would just say "oh, they were bribed to say it" or something.

@Anthony The usual annoying ping to see whether there's any sign of movement on this...

@Uncle Jeff Ugh.

1. The hospital he's at is a good one and so far as I can tell we don't yet have a terrible shortage of equipment or staff.

2. If someone is seriously ill, putting them on a long plane flight is not terribly good for their health.

3. At least some anti-inflammatories are believed to make COVID-19 worse.

Can we maybe skip the politicized cheerleading on this one?

@(darkives) Odds are nice because a Bayesian update-on-evidence just consists of multiplying the prior odds by the likelihood ratio. So, e.g., suppose you think Pr(something) = 0.4, and then you see new evidence E that's 10x more likely if something than if not-something. (Let's say it's certain if something and has probability 0.1 if not-something.) What should you now think Pr(something) is? In terms of probabilities: Pr(something|E) = Pr(something)Pr(E|something)/Pr(E) = 0.4 x 1 / (0.4x1 + 0.6x0.1) = 0.4/0.46 = 20/23. Kinda painful. (And I always ha...

This has nothing to do with predictions on this question, but I cannot get over the fact that there is a place in the US called "Gobbler's Knob".

The lower end of the range should possibly be exactly 2, since it's possible that someone will prove that there's an algorithm that runs in $O(n^{2+\epsilon})$ or something of the kind. (I agree with the judgement, implicit in the current range, that incremental improvements are more likely.)

aaaaaaaaaarrrrrrrrggggh

@Jgalt An astonishing outcome! Who could possibly have predicted that?

Carlsen just went 3-0 in the rapids and therefore wins the match. Question should therefore resolve positive.

So I took a moment to go past the FT's inconveniencewall and to my _astonishment_ (note: actual astonishment level: zero) the article does not in any way contradict the claim that Tesla sold a few cars at an (actual) price of $35k. It (1) reminds us that at one point Tesla's "after savings and incentives" price was just below $35k while the real price was above it (note: we all already knew this), (2) informs us that Tesla have tweaked their car-purchasing webpages to make the real price more prominent relative to the "after ..." price (note: this is a ...

@Tamay "At least one living person" sounds pretty good to me. Maybe specify that there has to be good reason to think that they were alive for, let's say, at least one minute after whatever they're in makes contact with the surface of Mars. (That means e.g. that resolution is negative if they crash-land and get completely smashed up, but positive if they land but the oxygen supply fails and they die a few minutes later.)

@KrisMoore I think that as a seasoned Metaculite you ought to know better than to be posting that sort of pure-noise comment. (I'm guessing that you're poking fun rather than seriously endorsing kimbob14's comment, but either way it's pure noise.) It's bad enough that we have this constant drip of new posters coming in just to say "yay Trump" or "yay Elon Musk" or whatever, but we could really do without actual established users doing the same.

I am aware that "bogus predictions" is a highly debatable way to describe the situation, and that if we take the resolution criterion seriously what we're predicting is as much the Metaculus community's behaviour as anything in the real world. But that itself is part of what horrifies me about this sort of question. Does the question title say "Will Metaculus predictors predict < 3% on this question?" No, it says "Will it turn out that Covid-19 originated in a research lab?". Without the recursive self-gaming bullshit element, this question could have b...

Yeah, I think this question urgently needs some clarity on what counts as a "transitional period".

The title says "... at the end of his term" which, at least at first glance, is ambiguous between "2024" and "whenever the hypothetical President Sanders leaves office" which could of course be either significantly sooner (death, impeachment, resignation) or up to four years later (if he wins again in 2024). The actual question text disambiguates this, but is there any reason why the title shouldn't be unambiguous too and say "in 2024"? I also suggest that the question should (explicitly) resolve ambiguous if Sanders is elected in 2020 but doesn't stay ...

The question asks not whether Tesla is fraudulent but what its market cap is going to be on a particular date. @alexanderpolta what "updating to the true state of the world" do you think might be called for? A company that has a large market cap due to fraud still has a large market cap.

(I am making no comment on the plausibility of the specific accusations here, not least because it's not at all clear to me what they are.)

User holahey has made exactly one prediction ever and posted four comments all of which were promoting his "Atlas Predict" thing. What's Metaculus's position on comment spam?

(I mean, obviously it's much more relevant than most comment spam -- it's not like these are comments on a literary-criticism blog saying "come and visit our porn site" -- but, still, spam.)

@Sylvain Project location radially onto earth's surface at the instant when it happens, obviously.

(Seriously: yes, of course, resolve ambiguous.)