I'm amazed every time I read/hear Nassim Taleb being quoted [1]. His black swan (outlier) "theory" [2] made some people rich, and it gained him fame.
He spent a lot of time pitching that "theory" (backing the least probable outcome) at economists at the likes of the ECB, BoE, the Fed, and so forth, and in the process pissing them all off very, very much.
Outliers are just that. Outliers. If, as an economist (or any other role that predicts anything), you went with a least-probable scenario you'd probably lose your job. So, instead, your recommendation is based on the most probable scenario. If you're thorough you might note outliers as a footnote.
[1] People who perpetuate and then capitalise on "bullshit baffles brains" annoy me.
[2] Predicting a one-in-a-billion probability that happens to come true is not a theory.
The outlier theory didn't 'make' anyone rich, it just explains why some outliers got rich and why other outlier events lead to the crash.
Taleb doesn't advocate backing low probability outcomes, because he can't predict which of them will occur any more than anyone else. That's a fundamental misreading of his approach. All he's saying is that some of those low probability events are going to happen, and you need to develop strategies that are survivable given low probability events. Strategies that assume no low probability events will (ever) happen are doomed.
The problem is, and he's very much aware of this, that strategies that are survivable against many low probability events are never optimal for the actual way events play out. The question then is what degree of non-optimisation (risk protection) are you willing to tolerate. A single trader might tolerate extreme risk for short periods of time. Too big to fail institutions should tolerate very low risks. Taleb is arguing for more realistic and practical ways to calculate and account for risks.
He spent a lot of time pitching that "theory" (backing the least probable outcome) at economists at the likes of the ECB, BoE, the Fed, and so forth, and in the process pissing them all off very, very much.
Outliers are just that. Outliers. If, as an economist (or any other role that predicts anything), you went with a least-probable scenario you'd probably lose your job. So, instead, your recommendation is based on the most probable scenario. If you're thorough you might note outliers as a footnote.
[1] People who perpetuate and then capitalise on "bullshit baffles brains" annoy me.
[2] Predicting a one-in-a-billion probability that happens to come true is not a theory.