“one of the best Covid forecasters there is” according to this tweet, which might be based on some quantitative metrics I haven’t easily found or might just not be right
There isn’t (ironically) a clean quantifiable metric for comparing forecasters [1] across different platforms and different question sets , but among the EA forecasting world (at least in 2020 when I paid attention to this), Juan has had nontrivial renown in having consistently one of the best coronavirus forecasting records across a broad range of platforms. For example, in addition to his native Metaculus, Juan currently is #3 for coronavirus questions among Good Judgement Open forecasters (and iirc used to be higher, note that he did not predict all questions). He was also #1 in a private GJP 2.0 tournament I was in. He became a certified superforecaster (TM) in the end of 2020. He also generally has explicit reasoning and good takes.
With the caveat that I mostly stopped paying attention to covid in 2021, I think Juan is plausibly one of the best covid forecasters out there, at least among people willing to put their forecasts out there publicly. I think it’s plausible that private entities (eg in intelligence agencies, or trading firms) have noticeably better private forecasts on covid than the public ones we’ve seen, but I’m pretty agnostic about this.
[1] The Rick and Morty joke “science is more art than science”comes to mind.
Re: your comment on Juan Cambeiro being
There isn’t (ironically) a clean quantifiable metric for comparing forecasters [1] across different platforms and different question sets , but among the EA forecasting world (at least in 2020 when I paid attention to this), Juan has had nontrivial renown in having consistently one of the best coronavirus forecasting records across a broad range of platforms. For example, in addition to his native Metaculus, Juan currently is #3 for coronavirus questions among Good Judgement Open forecasters (and iirc used to be higher, note that he did not predict all questions). He was also #1 in a private GJP 2.0 tournament I was in. He became a certified superforecaster (TM) in the end of 2020. He also generally has explicit reasoning and good takes.
With the caveat that I mostly stopped paying attention to covid in 2021, I think Juan is plausibly one of the best covid forecasters out there, at least among people willing to put their forecasts out there publicly. I think it’s plausible that private entities (eg in intelligence agencies, or trading firms) have noticeably better private forecasts on covid than the public ones we’ve seen, but I’m pretty agnostic about this.
[1] The Rick and Morty joke “science is more art than science”comes to mind.
This is helpful, thanks!