When The Bears Get Bullish

Sometime or the other the bears have to get bullish and the bulls must get bearish. That paradox has its roots in the inevitability of closing out positions sooner or later. A bull must unload and a bear must eventually cover its positions.

Thus although, the obvious initial impact of a big short sale or long purchase is to drive the scrip down and up respectively, the future impact is in the opposite direction. Thus the very existence of a large short or long position defines a stock being oversold or overbought. And when positions hit a certain level, the price of the scrip must switch its trend.

 

This is a good short term signal. On Wall Street, they define it as the short-interest and long-interest ratio. There is a further refinement available for traders on NYSE. Trades are demarcated as marked to general members of the exchange, specialists in the stock and general public. NYSE traders watch the specialist positions like vultures.

On the BSE, that level of demarcation isn't possible and the concept of specialist market-makers doesn't exist except loosely since jobbers tend to specialise. But the general impact of outstanding positions on a stock's price is by and large the same. It is likely to remain that way until and unless trading in individual futures is introduced.

The situation is similar on the NSE. The NSE doesn't allow rollover which means that it is quite easy to predict prices on the weekly settlement day. An oversold stock will rise and an overbought stock will fall as positions are cleared. The BSE allows long term rollover via the badla system. But even there the trend exists. But the timings are less easy to work out.

Here we outline a model which uses outstanding positions on the BSE to predict near term price reversals. It can be further refined by factoring badla rates but that would make it rather complicated.

The basic idea is simple. There are cumulative outstanding short and long positions in some 150 scrips. Each individual stock has its own trading patterns with an average level of outstanding positions.

When those positions hit certain levels the stock moves into a oversold/overbought zone. It can be expected to switch direction very rapidly. Traders and investors can take advantage of the reversals.

The outstanding positions can be confusing to read. But at any given session one will always exceed the other. And they can be expressed as a per centage of either average volume or of the daily volume of the stock. When they rise above certain levels they trigger trend reversal. We've attempted to work through three examples which may help readers refine this idea in their own studies.

The three scrips under scrutiny are HLL, ITC and Reliance. They have been studied for the past year. HLL has been bullish in that period moving from Rs 1445 to Rs 1725 to gain 19 per cent. ITC has been less bullish moving up 17 per cent from Rs 539 to Rs 634.

Reliance has been bearish dropping 26 per cent from Rs 183 to Rs 137. They are three of the biggest scrips in the Sensex and all highly liquid. However, only Reliance has mirrored the Sensex (down 26 per cent) in the same period.

We discovered that it was possible to set a simple ten per cent filter signal that seemed to work for the three stocks we scrutinised. If you take an opposing position every time one of the outstanding ratios climbs past the ten-per cent mark you ought to gain within three sessions.

Higher or lower filters could be set as people work out the probability of a quick reversal at any outstanding level. Also commissions and their impact would need to be factored. However difficult the finetuning may be, the primary indicator seems to work as well as a short term tool can be expected to.

HLL

HLL has comparatively low trading volumes given that it is the biggest stock in the market. It had an average daily volume of 1,56,000 shares. It has surprisingly high volumes of outstanding positions for a stock with a staid reputation.

It has usually logged larger outstanding volumes in comparison to total turnover. The short to long positions expressed as a per centage of average daily turnover is 5.22 : 8.05. As a per centage of actual daily volumes it is 8.5 to 9.19 respectively. Thus we see that generally higher long positions are positively correlated with long term rise in prices. There is a great degree of volatility and huge standard deviations. Cumulative short positions have ranged upto 76 per cent climbing above 10 per cent on 52 out of 205 trading sessions. Long positions have risen above 196 per cent of daily volumes but it has gone above 10 per cent only 40 times. Strong reversal signals tend to pop up whenever the scrip has a cumulative short or long position that is above 10 per cent.

What you get is several days of cumulative outstanding positions being up either in the shorts or the longs. Then this ratio tends to stabilise and the other position starts to rise. The price of the stock promptly reverses. It is worthwhile to start looking for a reverse position in HLL whenever the stock has an outstanding position of more than 10 per cent in either direction. Above 20 per cent the swing is almost certain to occur inside three sessions.

ITC

ITC is often the most volatile stock in the market where prices are concerned. It also logs very high volumes. But it has surprisingly low cumulative outstanding positions for a stock with such a wild reputation. It has average daily volumes of 4.07 lakh. However it averages short positions of only 2.92 per cent of average daily volumes. Long positions have been even lighter at 1.43 per cent. When comparing daily volumes with cumulative positions, the short and the long average is 3.49 per cent and 1.98 per cent respectively. This is at variance with HLL since the stock has risen over a year despite generally higher short positions.

Volatility in the levels of outstandings is also surprisingly low. In the last year ITC has logged outstanding short positions above 10 per cent only in 11 sessions. It has logged long positions above 10 per cent only 6 times. The highest the cumulative have reached is around 21 per cent (shorts) and 14 per cent long. The ten per cent signal filter appears to be very accurate. Almost like clockwork within three sessions of the ten per cent mark, prices swing sharply.

Reliance

The petrochem major has a reputation for being the perfect market mirror. It has large volumes and it has a huge number of individual investors so the scrip witnesses many small deals as well as large institutional deals. Among the Sensex scrips it is the most closely correlated to the Sensex itself. The stock is rumoured to be a beloved of traders and speculators.

But Reliance has lower levels of cumulative outstandings than HLL! The average daily volume of Reliance is 53,00,000. On an average the cumulative short position is around 4.74 per cent of that. As a per centage of daily volumes, shorts tend to be higher at around 5.88 per cent. Cumulative long positions are around 3.65 per cent of average daily volumes and 4.6 per cent of daily volumes. This shows that Reliance's bearish performance has been mirrored by generally higher shorts than longs.

Shorts have climbed to 35 per cent of daily volumes on occasion and they have been over 10 per cent on 38 sessions. Longs have been above 24 per cent several times and climbed past the ten per cent mark on only 26 occasions. Here again the ten-per cent clicks.

The stock usually reverses trend within three sessions of either cumulative position hitting 10 per cent.

Some of the findings were inconclusive or surprising. It doesn't appear that generally higher levels of outstanding positions means generally higher volatility. HLL is indisputably the least volatile of these three stocks. However it has higher levels of outstanding positions than the others. This is not translated into wilder price moves.

Also a long term bull or bearish trend may be accompanied by generally higher levels of either long or short positions. So there is no real predictive use in the long term in studying short-long positions nor do they necessarily impact heavily on volatility.

However as a short term tool, the concept of using outstanding positions as a direct overbought/ oversold indicator seems to work pretty well. With any individual stock the overbought /oversold levels will differ. However they can be worked out fairly rigourously.

And the correlation between high levels of cumulative outstandings and short term trend reversal is very high. The impact on short term trends is very obvious and this is probably a great tool for traders and investors who are hoping to time the market. Further refinements can be made by also taking general badla levels into account. If settlement considerations are also factored in, it maybe possible to eventually come up with a neural net/fuzzy logic sort of rulebook.