The common skill among the few great portfolio managers, investors, traders is dealing with uncertainty and the management of risk. The overwhelming majority is unable to obtain skill because they follow conventional wisdom – they are unable to think independently to develop better beliefs about the task. Conventional economics applies simple theories to explain complexity. They use linear equations like standard deviation in attempt to explain non-linear market data, even though they know it misses the outliers that are most important. If you intentionally ignore the potential for big moves and shocks (the fat tails on the ends of the bell curve), you’ll likely find yourself on the wrong side of them.

Mark Buchanan, a theoretical physicist, wrote a thought-provoking article for Bloomberg titled Mandelbrot Beats Economics in Fathoming Markets. He writes that market returns are non-normal, not the symmetry and equilibrium assumed by “Modern” portfolio theory. Mark references Benoit Mandelbrot, a name readers here will recognize as the author of The Misbehavior of Markets: A Fractal View of Financial Turbulence. Volatility and price shocks occur far more often than conventional economics expects, making its followers unprepared and anxious.

He makes some good points about the topic, one that I speak of often. Ironically, I also noted a logical contradiction that others may not notice. For example, he writes:

Larger movements of, say, 10 percent to 15 percent, are less likely than movements of 3 percent to 5 percent. And the probability of a movement decreases in simple inverse proportion to the cube of its size: If moves of 5 percent or more have a certain likelihood, then moves of 10 percent or more are 8 (2 cubed) times less likely, and moves of 20 percent or more are in turn 8 times even less likely. But they still occur with some regularity.

This is verifiable and true. You see, probability is a matter of historical frequency in the data. If there have been more movements of 3 percent to 5 percent, then they are more probable than moves of 10 percent to 15 percent that occur less often. If something has occurred much more often historically, you have a higher probability of experiencing it.

But, they have to be careful not to try to find simple patterns to explain everything. When I read some of the comments about such patterns, it seems they need to be careful not to find patterns just to make them feel more certain. Uncertainty from unknown outcomes is a thing to embrace from understanding it, not a thing you can avoid. We can't avoid unknown outcomes. The only way to experience unknown outcomes well (and be prepared to respond to them well) is to expect them and know they're going to happen. The paradox is that there is a fine line between gaining an edge from understanding the data and accepting uncertainty.

An early quote from the article: 

It seems that we’re complete suckers for the illusion of certainty and the seeming unlikelihood of the unthinkable, even though financial and economic history is one long string of crises.

If price trend data is asymmetric, you may consider that it may require Asymmetry™ to deal with it. That is, an asymmetric system with a focus on imbalancing risk and reward. 

You can read all the article HERE.

Hat tip to MM for pointing out this article for comment.

(Asymmetry™ is a registered trademark of Shell Capital Management, LLC)