Fooled by Randomness is a masterpiece written by Nicholas Nassim Taleb that truly elevated him to the “Guru” status. I first heard about this book when it was being used as a phrase when describing someone attributing a pattern to something that is plain co-incidence( ex:- Every time Sachin Tendulkar hits a century, India loses)
The book is by no means an easy read and I recommend it to people only if they are fascinated by what they find about the book on the net. So far I have read about half of the book and already I have come across many interesting concepts.
The main concept explained in the book is that every incident is based on probability. For everything that happens there could be 100 alternative outcomes that could have happened but didn’t. These other probabilities are present but hidden from people who don’t think in terms of probabilities.
In general, we believe that more knowledge we have about something, the better the decisions we can take about a particular situation.
He gives an example of a person asked to pick balls from a wooden box and come to the conclusion as to the contents of the box. This person can pick one ball( say red color) and assume that all balls are red. He might pick a second ball( say yellow) and assumes that they are distributed evenly( 50-50 red yellow). As he keeps picking the balls he can arrive at a better decision as to the contents of the box.
This is true in most of the mathematical problems we face.
However, in real life this is not true since the circumstances and situations change. Suppose you are driving on a smooth,wide, empty road. You drive for hours and yet you don’t see any potholes or any obstacle. Can you assume the road will remain like that till your destination? No. In this case you know the circumstances could change( you might enter a town)
Many investors in the stock market have been bankrupted because of these changed circumstances which resulted in a previously unrealised probability to become reality.
This is easily explained in terms of crows! No amount of black colored crows will prove that all crows are black. However, a single non-black crow can prove the statement “all crows are black” as false. To prove that all crows are black, one would have to take all the crows that exist on earth and make sure that all of them are indeed black.
[The author explained in this terms of swan. No number of white coloured swans can prove that all swans are white. This is well-known as Black Swan problem]
Nicholas knew that there are people like me and then introduces a new concept to bring be back to ground. He talks about skewedness probability. Would you be willing to play Russian roulette for 1 Million dollars? The chances of death are pretty high( 1 in 6). No right? Some of us would go ahead and try it but most of us would refuse.
Now, imagine a game in where 99 out of hundred times there is a chance that you will win 10$ for every 1000$ you put in. There is a 1 in hundred chance that you will lose all 1000$. Now, would you play the game?
This is a really hard question to answer, most of us would say yes, because in all our schooling we learn about even probability. That is schooling teaches that in a coin toss game you have a 50% chance of winning. the other 50% time you lose. So a person aware of probability will play the above game( Hey! 99% chances of winning!!).
However, In the above example, if a person plays 100 times, 99 times the person wins 990$, but the one time he loses, he loses 1000$ for a net effect of -10$. So, a person who is aware of probability will play, but a person who is aware of skewed probability will never play this game!
Even if you play the game, you should be aware that there is a 1 chance in hundred in which you might lose all your winnings and stop playing the game after some time. The problem with trading in the stock market is that the probability of losing everything would be something like 1 in 1 million or so. Since the probability of such an occurrence is so low, many traders forget that such an incident could happen and could ruin him.
The recent Banking meltdown is an example of this rare probability becoming a reality( The price of houses always goes up!).
Now, with this knowledge, would I drive fast on a smooth empty road? There could be a pothole deep enough to destroy my vehicle. I would drive a bit slower for sure…looking out for signs of bad road…but still not at 40 🙂
In the second part of the book, Nicholas explains about how there is a certain bias in all the conclusions made using historic data.
For example, in most of the management books I have read they take a set of good companies( based on performance for last 20-30 yrs) and analyse them. They then figure out things that are common to all these companies and then come to the conclusions that because of these qualities, the companies became good.
The author does not believe in any such conclusions as he says the analysts are confusing causality. He says that it is just plain luck that all the good companies had the same qualities. Causality cannot be established just by looking at a set of good companies because of Survivorship bias. That is, the author argues that there could have been bad companies that also had similar qualities and they could have failed in the past and hence didn’t show up in their sample list of companies.
The author says that only way to establish causality is to take into account ALL the companies that existed 20-30 yrs ago, analyse all of them and see what is common among companies that were considered good and check if the same qualities were MISSING in the companies that failed.
This totally destroys the argument of almost all the management books I have read and I believe Mr. Nicholas Nassim Taleb is right!
Management Concepts & Book Summaries 