Fooled by randomness?

by kirangarimella on September 24, 2013

Nassim Nicholas Taleb’s “Fooled by Randomness: The Hidden Role of Chance in Life and in the Markets is a highly readable book where the author, in an entertaining and frank style, explains how people are continually fooled by randomness which, according to him, affects and explains successes and failures to a greater degree than one would suspect.

For investors and traders, this has important implications.  On the one hand, we have the camp that insists that the markets are inherently random – the so-called Efficient Market Hypothesis (EMH), which comes in three flavors as it relates to tradable instruments:

  1. Weak EMH: all past, public information is already reflected in the price of the instruments.
  2. Semi-strong EMH: Weak EMH is true and all current, public information is already reflected in the price of the instruments.
  3. Strong EMH: Semi-strong EMH is true and all hidden (i.e., insider) information is already reflected in the price of the instruments.

In the other camp, we have traders and investors who ceaselessly create or follow systems that are built on the principle that the markets are predictable.

What do you believe?

I believe that EMH in any of its forms can’t possibly be true because even if all the information is publicly available, it doesn’t mean everyone has taken all of it into account and factored it in or discounted for it or remembered how they formed opinions in the past.

As far as trading is concerned, here’s my take: the markets are random, but what has that got to do with making money?

Huh?  You see, the markets are random, but they do have an underlying probability distribution.  Most people, when they think of randomness, assume that the underlying distribution is uniform.  For example, the distribution of heads and tails of an honest coin is uniform random, which means that the probability of heads (or tails) is 0.5.

No one said that markets have a uniform random distribution, but that’s what people assume.  They have a non-uniform distribution which no one can characterize (at least so far), but the distribution makes itself felt in the way the price charts show up.  If you don’t believe me, try building a random ohlc chart.  It will look like a price chart, but it will exhibit strange characteristics that an experienced chartist can spot as fake (for examples, a series of down bars where each bar has an open that is less than the close).

The markets are driven by the balance of fear, greed, and waffling with uncertainty.  This constantly shifting balance has its own smell (aka, the underlying probability distribution).

Even if the distribution were purely random, you could still make money at it.  Casinos do it all the time.  For example, let’s say you create a coin tossing game where every time it lands heads you get $2 from your opponent and if it lands tails you give your opponent $1.  You will come out ahead in the long run (assuming you can persuade someone to play this game with you – if you find such a person, do be a sport and share their contact info with me).

It’s called the edge.  If you can define a decent edge and stick to it with discipline, then you can make money.  Of course, making billions in 10 years starting from $1000 is attributable to good luck (aka, randomness).

Leave a Comment

Previous post:

Next post: