@honeypuppy Yeah, I was about to comment that this is trying to capture "when will Democracy take over China" but is actually much more likely to capture "when will China take over The Economist."
The main criticism of the Economist model seems to be that it's overconfident, with excessively thin tails. It seems to me that such a model will typically score a little better than a properly calibrated one, though losing out over time through occasional wild misses. I think this is correct.
That's right, I'm thinking that with careful planning I could GUARANTEE myself five or even SIX imaginary internet points in a mere 30 years hence
Trump hints that he "was all set" to call Bernie a Communist, until Biden won Super Tuesday:
I'll tell you, I was all set for Bernie because I thought it was going to happen. We get ready for things. So mentally I’m all set for Bernie, communist, I had everything down. I was all set and then we have this crazy thing that happened on Tuesday...
Trump +195,953 votes
6,277,135 votes reporting
~88% estimated vote reporting
~855,973 votes remaining
Biden must win remaining votes 525,963 to 330,010.
Biden must win remaining ballots 61% - 39%.
Biden is favored. Mail ballots are HEAVILY D.
Politico reports that "The Newsom recall just got real. The campaign to oust Newsom went from unlikely to unavoidable this week." Not sure why they thought it was unlikely, perhaps they should have been checking Metaculus!
@Reprisal I don't understand what this means except you didn't predict right.
I'd be interested in a short fuze question about pardons or commutations for one or more Capitol stormers over the next week-and-a-half.
Not looking great for the 97% Metaculus prediction here.
I think the underlying event is ~60% to happen but I'm not sure about whether the surveillance and reporting will actually make that clear until years afterwards. That was the pattern on a smaller scale with swine flu.
Since everyone hates too many questions about the coronavirus, let's have a perfectly innocent non-coronal question about whether final US Census apportionment numbers will be submitted to the President by year's end as required by law.
The current peak is in April which saw about 20 million confirmed cases worldwide, more than any other calendar month. Community right now has about 40% probability on impossible answers (January-March and May, unless a very surprising number of backdated confirmed cases comes in for May).