Bayes ‘ Theorem

Bayes’ Theorem Explained.

Its amazing how two different fields interact with each other and if you keep your mind open, you can apply axioms of science and mathematics on your daily life.

This is how Jacobi’s Invert analyses for maths can be used to solve any problem of life too, more on that later. To become an axiom or a rule or a law, a theory has to go through a set of very rigorous and stringent parameters. And then it has to prove itself in every situation. Presence of 1 black swan destroys the theory ‘all swans are white’.

That is the reason why technical analyses or value investing is not Science, whereas measuring the earthquake on a Richter scale is.

While Investing and life in general is not exact science, not even close, there is nothing better than solving the complex problems of life or investing through these time tested, proven methods of science and mathematics.

I came across Bayes theorem in mathematics and science section of my mental model development and I was just amazed to get an AHA!! when it directly correlated with my profession.

In laymen terms, let me explain BAYES ‘ Theorem (x).

What it essentially calculates is that how much probability of a given event should change in presence of changing (confirming or disapproving) evidence. Lets break it down further to bring this point home as I was given a feedback by a friend that i use difficult to understand lingo.

Bayes theorem calculates the change in probability when new information is added to the set. Lets say you think your company’s accountant Mr.X is a loyal, high on integrity fellow. You assign a probability of 90% to his not cheating you. Now, we add some more information 1) he has a criminal record. 2) he has 4 wives 3) he is in financial crises.

Bayes ‘ theorem simply calculates the changing probability from 90% to whatever with each set of information.

Lets use a mathematical example. There are 03 coins. 01 is a normal coin with 50% probability of heads and tails. 2nd coin is twisted and it is molded in such a way that it comes heads only 1/4th of the times. And the 3rd coin is heads on both sides (100% probability that heads will come)

Now you pick up a coin randomly. The odds that you have picked an all heads (both sides head) coin is 1:3 or 33.33% isn’t it. Its very simple so far.

You flip the coin thrice, and on all three occasions, it turned heads. What are the odds that the coin in question is the ALL heads coin.

According to the Bayes ‘ theorem . P(a|b) = P(b|a) P(A) / P(b)

the answer is 87.7% which is not important. What is important is that Bayes ‘ theorem teaches us to keep a track of evidence that shows up because it changes the probability matrix on its head.

It teaches us not to suffer from LIKING bias and CONFIRMING bias. Again, following my friend’s advise, NO high funda words. Let me give an example. What I am saying is that lets say you have a girlfriend and you obviously LIKE/Love her. And then you start getting these dis-confirming evidence that she ignores you, abuses you, treats you badly, nags you and confines your freedom.

Bayer’s theorem suggest with every new evidence, her probability of being Ms Right deteriorates and you should factor in these evidences and not discard them just because you like her.

Easier said then done. I know at least 30 people in a wrong relationship unable to say FUCK OFF. Similarly, I know a 100 cases where Investors ignored the dis-confirming evidence that something is seriously wrong with the company and are sitting with suzlon, renuka sugars, dlf, opto circuits and unitechs of the world.

The AHA moment for me using BAYES ‘theorem came when I realized that finding value stock is just half the battle won. I need confirming evidence of price moving up and promoter shareholding increasing to tilt the probability scale in my favor. Similarly, I don’t get married to the stocks I have bought. If the Price is falling below my stop/comfort level, I don’t hold on just because i think its valuable.

I had an interesting conversation the other day with one of the potential client who had some apprehension with my ModusOperandi. He said, I would love to join your subscription service however you are not hard core value investor and decide your exit based on price and not fundamentals.

That thought made me wonder that value investors (me included) are the biggest hypocrites in town. We want to find valuable company before any body else and then we want everybody else to find the same value. If they don’t find it valuable, my skill of finding value first has no meaning. We basically want to reach there first. In other words, we are openly saying that we are more intelligent, rational and smart than the average people who will join us later. As they say, idiots do what genius has done earlier. And then, we will remain with the company as long as the fundamentals are intact and we can measure that like no one else because we do not suffer from any biases and are RATIONAL super humans.

Now this way of thinking is correct and Warren Buffet’s way to get rich if you are actually ABOVE average. 90% of the people think they are, which is a mathematical joke. (how can it be average if 90% of people are above it)

Since I am not a genius like Warren Buffet with 20/20 foresight, I need to follow price as an indicator. Not the regular fluctuations of minutes, hours or days but WEEKS and MONTHS. But no two ways about it, I follow PRICE and I am proud of that. Price changes everything. Price moves first, fundamentals follow later.

If something has fallen from my weekly stop loss (which is quite big), and nothing is wrong fundamentally with my company. I will still get the hell out. Just because I don’t see it, doesn’t mean it doesn’t exist. (Availability Bias).

And I would like to use this blog as an opportunity to address all the 90% people who are above average to respect price. You are Not Warren Buffet, not even close so don’t copy him.