Black Swan Trading Systems

I have been going through my “drafts” folder this year trying to get more organized by finishing partially-written blog posts. Discussion in this post led to my thoughts on this from Sep 2018.

—————————

For me personally, I think the small sample size is still a
problem and it may be the reason why I’d be skeptical
that any statistically valid Black Swan system could
really be developed.

Ultimately, how could such a system ever be said to be
better than luck? Even if it hits the next one, two, or
three market crashes (over how many years or decades
would that take?), the sample size is still very small.

I think this is why I personally have never endeavored
to pursue development of such a system. This would also
be the root of my skepticism if I ever saw one.

I would offer up the caveat that from a practical
standpoint, hitting big on even one market crash could
make a boatload of money regardless of whether lucky
or not. I’m definitely not saying it’s a wasteful
pursuit: only one incapable of statistical validation.

Rather than developing such a directional system, I
wonder about having some long put insurance on
at all times. The insurance will lose a small amount
most of the time. Rarely, the insurance will pay
off big. Even if the total return is negative, lowering
the max drawdown of bullish strategies may still make
it a worthwhile addition (demonstrating this through
backtesting would probably face the same challenges
that I mentioned above, though). In addition, I could
always advertise “if the market goes to zero, the
system will profit handsomely.”