You are here

Short-Term Chart Warns We Could See Further Sell-off in Stocks

I follow a trading system that makes decisions on a weekly basis but can be applied to daily or monthly data. This week, the daily view turned bearish and could be warning of a sell-off in stocks.

My 26-week rate of change (ROC) trading system is designed to find the strongest ETFs in the market. It has a number of risk management tools built in because I believe that long-term investment success requires avoiding very large losses. I also define those risk management tools in advance because it is very difficult to make buy and sell decisions without rules. When relying on emotions rather than rules, I find that selling results from panic, and all too often, fear drives investors to sit on cash long after the selloff is over.

Some traders successfully use their emotional judgments to make trading decisions. But I believe most traders are not successful with a "fly by the seat of their pants" approach, and I know I do best with a rules-based system.

Another advantage of a system is that it allows for back-testing.

In back-testing, I have discovered that the 26-week ROC system could be adapted for daily or monthly use. On daily data, it offers a high degree of reliability, but there is a great deal of trading. I know from experience that frequent trading comes at a cost, and those costs are easy to underestimate. Commissions might look like a trivial expense, but there is also a spread (the difference between the bid and ask prices) and other factors that are difficult to quantify.

In Optimal Trading Strategies, Robert Kissell and Morton Glantz identify nine different factors that impact trading costs, and they conclude that professional investors incur a cost of about 1% on each trade they make.

When testing a system, I consider trading costs when analyzing the results. In general, I find that the cost of trading daily exceeds the benefits with a strategy like the 26-week ROC system, which is based on relative strength (RS). I also generally find that monthly action delivers about the same results as weekly trading.

In the end, weekly trading usually delivers the best performance when risk is considered because the sell signals are usually delivered faster than with a monthly system.

Looking at daily data, I do see a signal in a risk indicator for the 26-week ROC system that the stock market drop may have more to go on the downside. This is a warning that comes with SPDR S&P 500 (NYSE: SPY) only 2.9% below its all-time highs. It is important to remember that stocks are still close to all-time highs even after the market suffered a few big down days.

The chart above shows the results of the trading system (red line) with a 70-day moving average applied to those results (blue line). Signals are interpreted just like they are with any moving average system, with sell signals coming when the system results are below the moving average.

There is no action to take until the weekly chart confirms this signal. Daily data shows more whipsaw trades, which are trades that last only a short amount of time before being reversed. Whipsaw trades are one of the reasons trading costs are higher with daily data. They are not eliminated with weekly data, but they are less frequent.

Because we do not have any signals in the weekly data, the portfolio remains unchanged. Continue to watch stocks closely for signs of a change in the trend.