By Joe Krutsinger

Not all trading strategies work all the time. One way to decide which strategy to use is to paper trade them all and choose those that show the most promise, based on their equity curves.

Be honest. Do you believe enough in your current trading strategy that you’d stick to it come rain or come shine? Say you have four or five losers in a row -- will you continue to trade exactly the same way as before, or will you start bending your rules or perhaps even discard the particular approach completely?

The truth is that most traders abandon a strategy after three or four consecutive losing trades, thinking it no longer works. In reality, though, no approach will work well all the time. A trend-following strategy simply will not work in a choppy market, just as a top- and bottom-picking strategy is very unlikely to work in a trending market. It isn’t the strategy’s fault the market isn’t behaving in a way most suitable to its underlying logic.

The questions you need to ask yourself are, “When should I stop trading -- or re-optimize -- a strategy that is obviously not in tune with current market conditions? And, “How do I know when to start trading the same strategy again?” Alternatively, if you trade the same markets using several strategies, how do you know which strategies are currently the best to use?

Another side of the same dilemma occurs when you dig up that age-old strategy you once abandoned because it didn’t seem to be working, only to discover several years later that it would have worked like a charm had you just been patient enough and given it some time. But, there are tools that can help you determine when a trading approach is in sync with the market and when you’d be better off not trading it.

System Monitoring
A trader who lost more than $1 million trading pork belly futures was asked why he kept on trading the same strategy in the same market despite the obvious fact that it wasn’t working. His response: “This is the only thing I know how to do.” The trader truly believed that he had just been a little unlucky lately and that the only way to come back was to continue trading pork bellies using the same strategy he always had -- until his luck turned around once again.

It never occurred to the pork belly trader that it doesn’t matter which markets he trades or what strategy he uses, as long as he makes money. He should have cut his losses by changing either the market he was trading, his strategy, or both, and not fall back on the original market-strategy combination until the elements showed some solid proof of being in sync with each other again.

To avoid making the same mistake as the pork belly trader, you need to use a filter technique that will monitor your strategy’s performance from the inside, so to speak, and tell you when to trade or not trade a particular model on a particular market.

The JoeKrut Diff (JKD) and JoeKrut Measure (JKM) indicators are tools that enable you to monitor (but not trade) a system during its drawdown periods so that you can begin to trade it as soon as it shows signs of starting to perform well again. Paired together with any type of trading strategy, these two indicators can increase overall performance and strengthen your bottom line.

The JKM indicator is a 30-day moving average of your strategy’s equity curve. The idea is to trade a strategy only when the continuously paper-traded JKM is rising. For easier and quicker interpretations, the JKD indicator measures the difference in the JKM indicator from one day to the next. When the JKD is negative, the JKM is declining and, consequently, the strategy should not be traded in real time.

A Basic Strategy
To illustrate how these tools work, we’ll show how a trading system called JK-Call Buyer performs with and without them. The strategy we’ll look at has been tested over the full history of the NYSE futures, S&P 500 futures and the E-Mini futures with no changes made to the rules or the logic. So far, it has been more than 75 percent correct in all three markets.

The strategy works on end-of-day data and gives all trade signals as market orders the night prior to execution. Here it’s illustrated with the S&P 500 futures, but it can be used on individual stocks, as well as at-the-money call options. The strategy exits all trades after 10 days, regardless of any other factors.

The rules are as follows:
When you have no current position, enter long tomorrow at the market if:

• Today’s five-day relative strength index (RSI) is greater than yesterday’s five-day RSI 
• Today’s close is below the close of five days ago, AND
• Today’s close is less than or equal to the average of the last five days’ closes.Exit tomorrow at the market if:
• Today’s close is higher than the average of the last five days’ closes, OR
• You have been in the trade 10 days
  

Figure 1 shows the trade signals generated by applying this strategy on the S&P 500 futures from April 1999 through June 2000. Table 1 shows the strategy’s performance statistics.

Figure 1. A Short-Term System

 Table 1.Performance Summary

As you can see, there wasn’t much of a trend during this time period -- a long-term trend-following strategy likely would have lost money. However, the strategy had more than 70 percent profitable trades for a total profit of more than $75,000 and a maximum of two losing trades in a row.

But, this strategy, like any other, will under-perform from time to time. The next step is to see how the JKM and JKD indicators can improve the performance of this basic system.

Combining Strategy and Filters
By applying the JKD and JKM filters described earlier, you can avoid trading the strategy when it is out of sync with current market action.

Figure 2 shows the same strategy and market as Figure 1 but with the addition of the JKM indicator (the green line in the middle of the chart) and the JKD indicator (the red line at the bottom of the chart). Without changing the rules of the system, do this: When the green line in the middle chart is rising, trade the strategy; when the green line is falling, exit all real-time trades but continue to paper trade the strategy and track the hypothetical equity curve until it starts to move back up again, at which point you resume the real-time trading.

 Figure 2. Tracking Equity

For easier interpretation, you can look at the red line in the bottom chart, which will move into negative territory as soon as your strategy starts to under-perform and the JKM indicator starts to sink. In the case of this strategy, you can see that, over the last two-and-a-half years, there have been three major drawdown periods -- one that started immediately at inception of this test period, another that started when the equity had reached a total of $17,560 and a third that started when the total equity was $42,145.

The number at the bottom of each equity trough shows at what hypothetical equity level you would have resumed your real-time trading. Avoiding these three equity troughs would have improved your equity by $10,161. That is, instead of making $75,550, you would have made $85,711, while, in this case, shortening many of the losing trade sequences to one trade only. Admittedly, the time period covered and number of trades generated are not enough to generate statistically significant results, but the simplicity of the strategy should work as insurance against it breaking apart when applied to future, unseen data.

If you’re using several different strategies on the exact same market and the exact same time frame, you could simply look at the JKD indicator for each strategy and trade the one that has the highest value, which should be the one where the market and the strategy are the most in tune with each other.

*Reprinted (and modified) with permission