To Sell a Stock or Hold--When Is it Time?

Should you use the strategy of the long-term buy-and-hod investor or the short-term sell tactics of the trader in order to lock in small gains? Twenty years ago, it was relatively easy to categorize oneself as a trader or long-term investor. In recent years, the issues have been made more complex by the amplitude of market swings. Let us look at a few alternatives and possibly a strategy.

Imagine a straight line that rises 20% in a year. Assume that a stock is moving along this rising "trendline." Suppose the stock touches this line at the bottom left at point A, rises to a high at B about 12% above the line, declines back to the line and touches it at C, then repeats this pattern several times reaching peaks at D, F, and H and touching the rising line at E, G, and I (all during the same year). Of course this is an oversimplification. Peaks are not always 12% above a rising trendline, but that is not important to the point of this illustration. When the stock declines from point "B" to the trendline at point "C," there is no assurance that the stock will rise to point "D." Instead, support could break down at point "C." In a market like the one we are experiencing, such a breakdown could plunge the stock well below our purchase price within a single day.

If we bought with the intent of keeping the stock until it starts to break down, and the stock does not break down until it returns to the line of support at "I," the sale price would be the price at "I" minus the decline permitted by your stop loss. A common practice for investors who hold 15 or more stocks in their portfolio is to set their stop-loss at 15%, though this is far more than most traders would tolerate. Even so, the lesson is pertinent to both groups. Therefore, when the stock drops 15% below point "I," it is sold.

Now, what has happened? We bought at 31.625 and we sold at 36.70 for a gain of about 15.75% after commissions. That is not bad for little more than two months (we bought on 3/8/00 and sold on 5/23/00). Where did I get these prices and dates? This scenario is taken from Pfizer's chart pattern illustrated in Tutorial 1, and the reader should review it to get a more realistic idea about the probabilities involved. Nevertheless, the stock could have continued falling earlier at point "C," resulting in a loss. Next, assume that we decide to sell whenever the stock rises 10% above our purchase price (perhaps our rationale is that any time we can gain 10% within a three-month period, we are at that point annualizing over 40% and we should lock it in while we have the chance). In this case, we would buy at "A" for 31.625 and sell at 34.78. After we sell, the stock would climb to 35.25 before declining. We do not know how far the stock will drop, but we have taken a 10% profit off the table. It took only two days to get this 10% (we will net about 9.75% after paying brokerage commissions for buying and selling). We could buy again at "C" for 34.25 and sell at 37.675 for another net gain of 9.75% (the stock continues rising to $40). We might also buy a third time at "E" for 36.875 and sell at 40.56 for yet another 9.75% net gain (afterward, the stock rises to $44.25). Assume that because the stock did not touch the line at point "G," we invested in a different stock to get our fourth gain. By now, we have a net gain of about 43.9%. In the process, we have reduced our risk. Why? Because whenever money is still on the table it is exposed to potential loss. In this approach, we have taken money off the table part of the time, and we have locked in profits before the market has taken them away. Now, assume we buy again at point "I" and the stock drops 15% instead of rising, triggering our stop-loss just as it did under the other scenario. If we lose 15% on this trade, we will still have a net gain of 22% over the two-month period. Thus, we have greatly increased our net gain while reducing our exposure to risk.

The amount of gain set for our selling target would depend on the prior behavior patterns of the stock and on the stock's behavior during the time of our investment. The concept here is not that we would necessarily sell every position after a predetermined gain, but that in many cases we might sell preemptively.

Here is how I might rationalize a quick sale. I consider a return of 20% over a one-year time frame to be a fairly decent return for investors who ride out the ups and downs. For short-term traders, our target would probably be at least two or three times that amount. Imagine a stock rising smoothly in a straight line from $100 to $120 in a year. For the purposes of this discussion, consider this to be the model for our "ideal" stock. Now we overlay a plot of the growth of our real stock on this same chart. At some time our real stock may rise above the smooth line of ascent of our ideal stock. Whenever our real stock is above our 20% target line, our real stock is giving us a higher annualized rate of return than our "ideal" stock (that's 10% in less than 6 months, or 5% in less than 3 months). It's at times like this that I would consider locking in the higher growth rate and moving on to another situation. My reason is that locking in 10% in perhaps a quarter is far more appealing than risking that gain by holding for another three-quarters of a year for the possibility of getting an additional 10%. However, I would not necessarily sell immediately. I might be inclined to place a stop-loss just under the stock. If it continues to rise, I would follow it up with a close stop. Eventually, the stock will "twitch" the wrong way and I will be out of the position with a very nice return relative to the time invested. Of course, we could set our targeted growth rate at something other than 20%. Wherever we set it, the procedure would be the same.

Why sell just because the rate of gain has surpassed some targeted rate? Because periods of price acceleration are usually followed by an unwinding of the acceleration's gain. On the other hand, when a stock returns to its rising trendline, it is much more likely to surge in price. When the stock begins to bounce off this trendline, a trader may infer that the stock "wants" to rise. That is what a "setup" is. It is a pattern of behavior that often precedes a surge in price. It is all about weighing the probability of a gain from the current price against the probability of a decline. When we have this kind of configuration, the probabilities are aligned in our favor. Once the stock has moved and achieved a nice gain relative to the holding period, the probabilities are no longer aligned in our favor. Instead, they favor a decline.

Copyright 2008, by Stock Disciplines, LLC.