How accurate are Predictions

In markets, everyone loves to predict. Technical Analysis has quite a few tools / methods whose whole concept is the ability to forecast given certain circumstances. In a way, ability to predict is like the ability to act like God and is very very addictive. Its one thing to have a opinion of the future and quite another to predict that markets shall do this or that by this or that particular date.

In a google group called LongTermInvestors, there is this guy by name Ravi Palwankar who has gained a following based on his ability to give a date on which he expects the markets to move. The only caveat is that he does not provide the direction and hence it can either be bullish or bearish in nature.

His messages are like the one below (his most recent one)

I came across this guy only due to following Vish on Twitter who not only provides updates as and when a new mail is posted but also has kept records of his previous predictions. Thanks to him, I could make an attempt at testing the profitability of his calls.

Since the move can be either Bullish or Bearish and also can be + or – 1 day from the day he has given, I have in the test bought a Straddle (closest strike) 2 days before the event and sold 2 days after the event. In the case, the D-2 day falls on a holiday, I have taken the previous day closing price as entry and in case the D+2 day falls on a holiday, closing price of the next trading day is taken for the calculation.

Results are as here under

The sample size is pretty small and hence do take the results with a pound of Salt. But for what they are worth, it seems that shoring a straddle before the event has a better outcome than buying one.