We need to treat forecasting performance as a function of time. A strong majority of resolved binary AI questions had less than a year between their publish date and resolve date. (The same is likely true of continuous questions, though I didn't check.) But some of the most prominent open questions are not expected to resolve for several years. So the question is, how well does Metaculus predict the far future? Common sense tells us that forecasting performance drops off pretty quickly as the time until resolution increases.
If I'm reading this correctly, the 2020 delisting mentioned in the question text has been vacated by a district court. Still no change at the url in the resolution criteria. @samuelraasch do you have a take on this?
Temperature index for June is here. Lower than the last two years.
@Rexracer63 Apparently there's a new episode today. There doesn't seem to be any ambiguity here. It should resolve negatively.
As of August 13, FiveThirtyEight is at 50.0%. (If you download their data, their estimate is actually 49.966687.) In their latest podcast, Nate blames the delta variant. Quinnipiac polling from early August finds (PDF) that approval has fallen for Biden's handling of the virus and the economy, but risen for his handling of foreign policy.
— edited by Ab5A8bd20V
The Wikipedia page has been updated. As far as I'm concerned, the opinion of the admins is the only remaining source of uncertainty here.
If Kruger is 33%, and KBJ is 62%, then Childs is 5% at most. Seems low, no?
Evidently the UNSC is meeting on Sunday related to this issue: https://www.axios.com/united-nations-security…
Ukraine's member has commented: https://twitter.com/SergiyKyslytsya/status/14…
If the Security Council invokes 377, this question would not resolve positively, correct?