Comment by beeskel on The EMH Aten't Dead · 2020-05-16T09:52:42.083Z · LW · GW

Something that occured to me while reading this post:

In a given situation, some people who are not at the bleeding edge of finance in general will be well placed to beat the market in a way we would tend to think of as due to skill rather than luck.

In the coinflip example, the people who correctly predict the sequence do so through luck. We call the sequence random because we are unable to model the process which generates it, and we call it luck when they are able to predict it because they don't know how to model it either. In the case of detecting a nascent pandemic however, we do know some things about what makes you better able to see it coming.

If you are attuned to global risks, plugged into discussion networks that provide early warning of interesting developments in the area, have got the probablistic thinking thing down, are the sort of person to hoover up enough details to distinguish between a Spanish Flu-like scenario and a SARS-like scenario, and are enough of a disagreeable contrarian to not worry what anybody else thinks until you have vetted their sanity thoroughly, then you are a member of a subpopulation that is especially well placed to spot the shape of things to come in the particular case of covid-19. Add in previously aquired knowledge about how to make the relevant trades, plus a lack of laziness, plus getting struck by a little inspiration that doesen't come to everybody, and you are Wei Dai or Jacob instead of me.

How do you end up in that subpopluation? You do so randomly, if we think of it as drawing that particular ticket in the birth lottery. So in one sense you randomly got lucky, but in another sense you succeded through skill.

I don't know how much of the apparent discrepancy between the EMH and the fact that some people timed the downturn is explained by this, but I think it explains at least part of it.