This is by a large margin the most exciting improvement on Metaculus in a very long time. Thank you!

I predicted substantially above @Jotto's 33% figure, _after_ reading Jotto's comments. My reasoning is as follows. 1. Let's begin with Jotto's ~20% baseline. It would be an oversimplification to say that this tells us how likely a recession is without something like Brexit, because several of the recessions that feed into the baseline probability were themselves the results of unusual economic shocks. Let's say a 10% chance of recession if somehow Brexit is avoided (and the avoidance doesn't itself cause economic disaster via, e.g., social unrest from ...

@Tamay I can write such a bot right now. All it has to do is search for a question that mentions Elon Musk or Donald Trump, make an inane rah-rah or boo-boo comment, and then never interact with Metaculus again.

I suggest adding the not-having-tachyons feature.

Surely this should resolve now. Ambiguously if it must. FWIW, I think that ... If this is about the EP: * It should resolve positively if RE is considered "the same" as ALDE for this purpose, negatively or ambiguously (depending on how not-the-same) if not. If it's about the EC (I still cannot imagine how the question can reasonably be read as saying that, but frxtz says that's the intention and it's frxtz's question): * It should resolve positively if RE is considered "the same" as ALDE for this purpose and having 23/26 Commissioners belonging t...

Donald Tusk on Twitter:

The EU27 has agreed that it will accept the UK's request for a #Brexit flextension until 31 January 2020. The decision is expected to be formalised through a written procedure.

I think that's sufficient for this to resolve negative.

On the Stack Exchange network of Q&A sites, questions that seem to be attracting a lot of "noise" can be protected, meaning that only people with a certain number of Meaningless Internet Points are allowed to post answers to them.

Perhaps we should have a similar feature for questions that attract an exceptionally high level of nonsense comments and extreme predictions. (Looking at you, SpaceX On Mars question...)

There are already rankings for the last 30 days, 3 months, 6 months, and 12 months. It seems to me like that does pretty much the same as having "seasons" would. And I'm not sure how much seasons would really help. I usually sit somewhere around rank 5-10 on _all_ of those. Travis is usually at #1 on all of them. The shorter-term ones are noisier, but the very top-ranking people aren't just _older_ than the rest of us, they're also _better_. If you introduced year-long seasons, Travis would still be winning them all. Obviously the case gets stronger if...
Possibly relevant: [New IRA says Brexit has provided it with opportunity](https://www.reuters.com/article/uk-nireland-violence/new-ira-says-brexit-has-provided-it-with-opportunity-sunday-times-idUSKCN1S4079). I remark that there's something odd about this question's resolution conditions as they now stand. The _title_ says "If the UK leaves the EU, will there be a resurgence of the Troubles?". The _condition_ doesn't mention the UK leaving the EU at all, except for a rider at the end saying that the question resolves ambiguous if the UK isn't out by 202...
Resolve negatively because * question title says "national" * question text says "in Wikipedia's list of national chains". Resolve positively because * Vox prediction on which this is based doesn't explicitly say "national" * Link therein to an article about Impossible's own prediction (which is what Kelsey Piper says she expects to be met) doesn't say or imply national Resolve ambiguously because * the above lists are both nonempty I personally will lose MIPs if this resolves negative but think it probably should do. I _also_ think that when a que...

@Uncle Jeff, the only big protests for which I've known anyone who went -- this is in the UK rather than France or the US -- showed no obvious sign of being "bought" by any big sponsor. What is your evidence that the Gilet Jaune protests have been manufactured in such a way?

Your "track record" only shows binary questions. There should be some way to include numeric questions in it, or a separate thing for numeric questions, or _something_. Rationale #1: They're part of what you've done. Obviously they should be in there. Rationale #2: As we all know, the purpose of Metaculus is to hoard those precious preciouss pointssssses, and for me at least range questions are very often the ones that yield the most of those. Rationale #3: If you're going to look at someone _else's_ track record, using (_spit_) tachyons (note: actual...
I think these are good suggestions, assuming that the goal is to pick questions that advantage fourth-graders relative to today's text-predicting AI systems. To make some things more explicit: * GPT-3 is not good with _letters_, partly because it never really sees them; its input is encoded [in a different way](https://towardsdatascience.com/byte-pair-encoding-the-dark-horse-of-modern-nlp-eb36c7df4f10) that makes it easier for it to learn word-level patterns but [harder for it to learn letter-level patterns](https://www.gwern.net/GPT-3#bpes). I expect t...

@pranomostro At American universities, "class of X" means the people expected to graduate in year X. (More precisely, it may mean either the ones who actually do end up graduating in year X, or the ones who at the time of admission are expected to graduate in year X. For most people there's no difference between the two, but if you skip a year or defer a year or something then they differ.)

The title says "get the required validated signatures and trigger a recall election". The resolution condition in the question only concerns the first part of that. What's the intended resolution if one of these petitions collects enough validated signatures but for some reason a recall election then doesn't happen? (E.g., the rules change, or there's some sort of legal challenge, or the administration decides to go for blatant corruption and just doesn't do what it's meant to, etc.)

@prjdnath I think you have misunderstood the point of Metaculus, which is to try to make accurate predictions, not to say what you want to happen nor to declare which side you're on.

This is a longstanding complaint. I forget whether there's an actual good reason for the stupid-seeming thing the Metaculus scoring system does when a question resolves early, but what's certainly true is that the issue has come up several times before, people have very reasonably complained about it several times before, and nothing has been changed :-).

@TamayBesiroglu That's very nice, but e.g. North Korea claims to be a democracy already so I'm not sure how much stock one can put in such claims...

Coming in late here to agree with everyone else: leave it as it is.