Of Helmets & Stop Losses

Bangalore (like many other cities) has had a rule for compulsory wearing of helmets by those riding a two wheeler. But what I observe is that many a helmet are worn to protect not the head but the risk of getting caught by the police constable. At the first instance of a risk to the head, the helmet would be sure to bail itself out leaving the head to take care of the mess by itself. And since the wearer of the helmet in a way thinks that his head is protected, he may actually take more risks than what he would have when he was not wearing one (before it was made compulsory).

So, what is the relationship between the helmet and the stop loss you may ask. Well, there is one. A wrong stop loss like a wrong helmet may cause more grief to the user than to a trader who trades without one.

A stop loss is said to protect one from the loss while allowing the winners to remain in the system. But the thing about stops is that if not placed right, it has a ability to cause more damage than one where it was not used in the first instance.

So, what is the ideal stop you may wonder. Is placing a stop above a resistance / below a support a good idea? Or should one just stick to one’s risk profile and say that if a stock loses more than X% of my capital, I am out?

With the increase in algorithmic trading, stops below support / resistance are the easiest to take out since everyone sees more or less the same chart and comes more or less to the the same conclusions.

As to those using X% of risk per trade, the problem comes by the way of the fact that the stock may not necessarily fit that profile. Some stocks have very high volatility while some remain bland for most of the time but then spurt up in one swing what would be a multi X deviation from the mean.

So, when is that one should sell or buy (based on one’s existing long or short position). To me, the best way is the way described by Jesse Livermore (which I shall paraphrase) where he says that if a stock is good enough to sell, it should be good enough to short as well. So, if you are placing a stop at level X, not only should you be happy to sell your longs there, but also be willing to go short.

That does not mean that one has to go short every time one wants to exit a long, but its the conviction that matters.

On the other hand, if you are using a portfolio / ranking based model, the above assumption may not be required since you will be exiting one stock in favor of other which your model is showcasing as one with a better opportunity.