The art of trading

MARKET INSIGHT

How much can you rely on the laws of chance? A coin-toss has equal chances of heads or tails. However, as cricket fans know, there can be long consecutive sequences of heads (tails) and knowing the formula doesn't make it possible to predict the next drop.

Stock market fluctuations are far more complex and random. They cannot be accurately modeled. A stock moves in three possible directions "� up, down, or sideways. So far as a trader is concerned, he loses on two out of these three movements.

The moves are not equally likely. A sideways move (no change, or change lower than brokerage payable per trade) is least likely. Over the long-term, there is an upwards bias. But there are also long downtrends. Every move has different magnitudes. Frequency distributions of price change have been mapped. But no formula exists to predict direction or magnitude of change.

Successful trading starts with a mental estimate of maximum possible gain or loss. Contrary to popular opinion, good traders are not much better than amateurs at divining price direction. But they are much better at maximising gains and minimising losses. There are certain techniques that all good traders appear to have in common. One is the stop-loss.

Before entering a position, the trader will decide how much he can afford to lose and set a stop. Even more importantly, if the stop is hit, he will have the discipline to cut his losses. An extension of the stop-loss is the trailing stop. This is designed to lock in profits in a winning position. Once a position goes into the black, the trader moves up the initial stop. He will reiterate this while the stock trends up. When the trend reverses, the trader books a profit.

A third technique most good traders seem to follow is pyramiding. In a profit-making position, they commit more resources. That way, if the trend holds, the absolute returns are higher (though the return on capital employed drops).

If strictly followed, these two-and-a-half techniques cure the most common trading errors. The biggest losses come when a trader averages down in a losing position rather than cutting out when a stop is hit. The discipline of pyramiding only winning positions ensures good money doesn't disappear chasing the bad.

The other big mistake traders make is mistiming profit-booking. The nervous sell too early; the optimistic hang on until long after the trend reverses. A trailing stop takes care of both tendencies.

However, where does one set stops? If you're trading a stock with an assigned limit, it's easy. You know the maximum swing in a given session. So, you can calculate stops in terms of the number of maximum losing sessions you are prepared to absorb.

If you're trading F&O stocks, there is no mechanical limit. You can try the value-at-risk formula, which exchanges use to calculate margins. But there is a danger to VAR, which is well articulated by Nassim Nicholas Taleb.

VAR assumes price-changes have normal distributions. This is not quite true. A normal distribution assumes price-change will be within one standard deviation of average about 68 per cent of the time. It will be within average plus/minus two standard deviations about 95 per cent of the time and within three standard deviations 99 per cent of the time.

Given roughly 250 trading sessions per annum, even normal distributions suggest about three annual sessions of extra volatility. With volatile stocks, the standard deviation may be greater than the average but that isn't the worst problem. The issue is that moves of average +/- 3 SD occur much more often "� price volatility distributions have long tails. Such swings can be magnitudes larger.

Take the Nifty for example. Since January 2000, in 1935 sessions, it has an average daily price change of 2.04 per cent with a standard deviation of 1.34 per cent. That means a session with a price swing of 6.06 per cent should not occur more than 18-19 times.

In reality, the Nifty has registered 35 moves of greater magnitude! It has registered double-digit moves twice and 9 per cent moves on three more occasions. If you're on the wrong side of that in a margin position, you're dead and buried. Managing stops in such scenarios is a gray area where trading becomes an art rather than a science.