Trading 2010: Testing what worked

During the last one, three, and five years, the market has taken many traders and investors on a roller-coaster ride, providing many opportunities to make or lose money. As the year ends, it can be fascinating and educational to look back at how various trading strategies performed.

One powerful method that many traders use to gain an edge is backtesting. Backtesting allows traders to gain insights into which of the hundreds of strategies worked or fell short. By using the basic tools in Fidelity’s Wealth-Lab Pro,® you can use these strategies as a starting point—before experimenting, tweaking, or creating your trading system. Of course, while understanding what happened in the past can be educational, it does not guarantee future results.

The top strategies: Making the most of volatility

Fidelity’s Kent Thacker, director of brokerage products, backtested all of the 30 trading strategies that come preprogrammed into Wealth-Lab Pro® against the stocks in the S&P 500® Index over the last one, three, and five years through October 31, 2010. For this test, he assumed an investment of $10,000 per trade, no stop losses, and included commissions.

The rocky road of the last few years provided lots of opportunities, and the volatility made winners of some trading strategies, but created challenges for others. In the past year, countertrend strategies were top performers. These strategies are designed to recognize when a significant downtrend has occurred and attempt to profit as the stock moves back up. All three of the top performing strategies in the 12 months through October 31 were countertrend strategies: Neo Master, CMO Signals, and RSI Agita (see below for more detailed explanations of the strategies).

Over a longer term, a number of other strategies did well, including several dip buying strategies—which attempt to get into a stock after a sharp decline and profit on the pullback. For the three- and five-year periods, RSI, LLS, Dip Buyer and 3×2 system took top spots based on net profit.

Interestingly, in a similar study conducted at the end of 2009, the Moving Average Crossover Strategy was a top performer on a five-year basis, along with the RSI Agita and LDL2 systems. The increased volatility has created whip-saws for Moving Average Crossover investors. That has helped to move its performance down into the middle of the pack among the tested strategies.

“This kind of a strategy actually has only about a 40% win rate for the trades, meaning the majority of trades don’t work out. That win rate is caused by false breakouts where the cross could happen and then cross right back under,” says Thacker. “So you get quite a few smaller losses and some large losses, but the winners have a much larger average size—that helps the performance. But increased volatility could reduce that number of winning trades and reduce the size of the average profit.”

The hypothetical backtest results:

2010-1

Past performance is no guarantee of future results. All results for illustrative purposes only. Results according to WealthLab Pro,® based on $10,000 trades for the 12, 36, and 60 months through October 31, 2010 using all the component stocks in the S&P 500 Index.

2010-2

Past performance is no guarantee of future results. All results for illustrative purposes only. Results according to WealthLab Pro,® based on $10,000 trades for the 12, 36, and 60 months through October 31, 2010 using all the component stocks in the S&P 500 Index.

2010-3

Past performance is no guarantee of future results. All results for illustrative purposes only. Results according to WealthLab Pro,® based on $10,000 trades for the 12, 36, and 60 months through October 31, 2010 using all the component stocks in the S&P 500 Index.

Understanding the categories

Before we analyze the results, let’s look at the most important categories. First, net profit is the total amount of money the strategy hypothetically made during the time period. Although the strategies were ranked by net profit, there are other factors to consider before choosing a trading strategy.

For example, many traders believe winning percentage is extremely important. Strategies with high winning percentages will draw down less of your capital, and are considered easier to follow, because there are fewer losses. Another important category is the total number of trades. No matter how high the net profit, if the number of trades is excessive, not only is it time consuming, but it can generate a lot of commissions.

Another important category to consider is maximum drawdown, which is simply the point where you take your greatest loss.

Professional trader Pascal Willain, author of Value in Time: Better Trading Through Effective Volume (Wiley, 2008), explains how to use drawdown: “It’s a personal decision how much blood you’re ready to put in the fight,” he quips. “The maximum drawdown for my trading position is usually 8%, but on strong stocks that I’m confident about, I could allow 20% drawdown. If the drawdown is too high, I usually avoid that strategy.” He does point out, however, that a stop loss will usually reduce a high drawdown. “If you’re still losing money even with a stop loss, then there could be a problem with the strategy.”

In the above table, the date of the maximum drawdown for the three- and five-year period was November 20, 2008, when the Dow Jones Industrial Average (DJIA) closed at 7,552. If the drawdown is unusually high, it’s suggested you reevaluate the strategy, or use stop losses. More than likely, most people cannot realistically stick with a strategy that is losing more than 20%.

Understanding the strategies

1. Relative Strength Index (RSI) Agita

Strategy: This strategy includes the word, Agita, which comes from the Latin word agitare, or to stir up. One of the goals of this strategy is to take your profits early, before you get too anxious. Normally, when RSI, using a 14-day time period, reaches 70 (overbought), you think about selling. When RSI reaches 30 (oversold), you consider buying. This strategy was designed to take profits early, at the 55 level, just as it crosses the centerline, rather than at the 70 level.

Analysis: This strategy customizes the RSI parameters. The lesson: When using a backtesting program, it’s recommended that you experiment with different parameters until you find one that works for you. This strategy forces you to take profits early, and RSI Agita soundly beat buy-and-hold during the three- and five-year periods.
2. LDL2

Strategy: This is a technical trading strategy that attempts to buy stocks that have dipped. The system tries to enter at oversold levels and exit when the overreaction has subsided and the stock has leveled off. To calculate an oversold condition, the strategy sets an entry limit price by adding together the low and the close of the current bar, dividing it by 2, and then multiplying that by .94 to get the limit price. Once a position is established, it is sold after two days.

Analysis: Although LDL2 performed well in the three- and five-year backtests, it didn’t do as well during the last year. One possible reason is that this strategy outperforms in extreme conditions. While the market has been rocky at times during the past year, the overall trend has been up. One complaint about this strategy is the excessive number of trades you have to make. Nevertheless, LDL2 beat buy-and-hold during the three- and five-year periods, but not over the past year.
3. Neo Master

Strategy: Neo Master uses the Chande Momentum Oscillator, a range-bound (from +100 to –100) technical indicator that attempts to capture stock momentum. Neo Master buys after nine consecutive bars (a bar equals one day) where the closing price is less than it was four bars ago.

Analysis: Although Neo Master ranked high in our backtest, you’ll end up trading each symbol on the S&P 500 at least three times, which some may consider excessive. Nevertheless, Neo Master beat buy-and-hold during a three-year period, but didn’t outperform over the one- and five-year periods.
4. CMO Signals

Strategy: The CMO, like the Neo Master, uses the Chande Momentum Oscillator. Using this strategy, the system tries to enter when prices are lower, and sells when the oscillator says the stock is overbought or hits a 10% profit target, whichever comes first.

Analysis: Similar to a buy-on-the-dip strategy, CMO Signals is designed to buy when stocks are lower. It also sells when the stocks are overbought and emotions are extreme, helping traders to avoid getting greedy.
5. Dip Buyer

Strategy: Dip Buyer is one of the simplest strategies in Wealth-Lab Pro. You buy when a stock goes down by 8%, and sell it after holding it for one day. If, for example, a stock plunges by more than 8% at the open, you’d buy and hold for one day.

Analysis: Dip Buyer, like CMO Signals, attempts to buy overreaction and sell quickly, hoping to capture a bounce-back. Although it’s not the highest performer, its low drawdown and high winning percentage may make this an attractive strategy. “Sometimes the simple strategies work the best,” Thacker points out.

2010-42010-5

An intriguing trading strategy: Gap Filler

Strategy: With this strategy, you buy the day after there is a gap of 8% or more during the day (not at the opening). You then hold the position until the gap is filled.

Analysis: Sometimes, looks can be deceiving. Although not included in the top five, the Gap Filler strategy has a lot of potential. Why? When we studied the results, Gap Filler had the highest winning percentage (86%), one of the lowest drawdowns, and a very low number of trades (645) when compared with other strategies.

Although Gap Filler performed well, the downside is that there are no guarantees the stock will snap back to fill the gap. Therefore, you could be holding an open position for months, if not years. Nevertheless, traders who want to make fewer trades and end with higher winning percentages might want to take a closer look at this promising strategy.

The Bandwagon Strategy: Still in the basement

Strategy: Using Average True Range (ATR) as its main indicator, Bandwagon attempts to measure a security’s volatility over a certain period. The system enters the position, long or short, and instantly calculates an exit point.

Analysis: For the second year, the Bandwagon strategy was the least successful strategy for the three- and five-year periods, although it did perform a little better during the past year. Still, the buy-and-hold strategy outperformed the Bandwagon during the previous one, three, and five years. If you use a strategy that ends up in the bottom, it’s recommended you learn why. While many strategies have different merits, it’s important to understand how they react to different markets: A strategy that disappoints during one market cycle may outperform in another.

Backtesting basics
Although backtesting can’t guarantee future profits, it’s similar to finding the starting point on a roadmap. That is why many traders believe so strongly in backtesting. For example, Willain does overall market backtests, sector backtests, and stock-by-stock backtests. “Backtesting is extremely useful,” he says. “It allows you to see how your strategy worked in the past, and whether you get a positive probability.” The challenge, Willain admits, is finding an environment that is similar to the current market situation. “If it’s not the same environment, the possibility of success could be lower.”

If you learn that a strategy performs well on a backtest, you can use this information to investigate further. It’s also recommended that you work with a group of strategies than just one or two. Trader and author Toni Turner also suggests: “No matter what rule-based strategy you choose, the strategy should be combined with your tolerance for risk, the goals and objectives you wish to accomplish, and a sensible management plan.”

Share on FacebookTweet about this on TwitterShare on Google+Share on LinkedInEmail this to someone