From the December 01, 2010 issue of Futures Magazine • Subscribe!

The art of war and the science of trading

The strategy

The setup finally accepted, after several months of analysis, was fading (trading in the opposite direction) the opening gap in the equity indexes. Gaps occur daily and historically more than 70% fill, or retrace to the prior day’s closing price, by the end of the day. The entry and exits are easily predefined: simply enter at the opening bell and exit at the prior day’s closing price once the gap is filled. If the gap doesn’t fill by end of the day, close the trade and sleep well without the fear of overnight surprises.

But there is a critical step: estimating profit expectancy. The formula is:

[(Probability of Winning)*
(Expected Profit)] – [(Probability 
of Losing)*(Expected Loss)]

Traditional discretionary technical analysis suggests that stops should vary for trades based on the differing levels of support and resistance suggested by current market conditions. However, you cannot use the above formula to estimate your odds of a winning trade, nor how much you expect to lose, if you don’t know exactly how wide or narrow a stop you use. Without the ability to backtest variable, chart-based stop placements, it was decided to backtest the past 10 years of gaps in the Dow Jones Industrial Average and use whatever had worked best. The key number was about one-half of 1% of the value of the index.

This stop width permitted the fading of the opening gap, and accommodated the normal post-open volatility. It captured the majority (about 85%) of all gaps that filled, without first being stopped. It also limited losses on those days that the gap did not fill but instead continued in the direction of the opening.

Most important, a fixed and known stop size allowed the calculation of profit expectancy. That was the good news. The bad news was that it only generated a modest profit historically after commissions and slippage.

Improving the technique

Sun Tzu delivered another gem with this quote:

"He who knows when he can fight and when he cannot will be victorious."

To make money trading a fixed stop and pre-defined target (gap fill), you must know when to fade the gap and when not to fade the gap. In other words, our gap system needs to be optimized to select only the best risk/reward opportunities. Specifically, it’s necessary to calculate when conditions favor a gap fill with low opening volatility, and when they do not.

A simple, but effective technique for analyzing and selecting the best gap fade opportunities is using the opening location or "zone." By using the open, high, low and closing price of the prior day and the direction of the prior day’s open-to-close price action (up or down), all historical opening gaps could be segmented into one of 10 zones — each with significantly different gap fill win rates (see "Gap zone map," below).


While additional filters, such as market conditions, candlestick patterns and seasonality (day of week and part of month), can eke out more gains or better risk control, the core of the strategy is robust. It performs in all types of markets.

A challenge with discretionary strategies is producing a reliable back test. However, this mechanical gap trading technique back-tests well and performs profitably with real trading too. The actual equity curve for 2008-2009, trading 10 lots in the e-mini S&P 500, is shown (see "Equity curve" on below).


Depending on your temperament and risk tolerance, focusing on the opening gap can allow you to avoid many of the issues some traders struggle with while swing trading. Second, while many strategies will work in the markets, the key to success is finding what works for you — in this case, planning to win the fight before pulling the trigger and making sure the gun is aimed in the right direction.

"The general who wins the battle makes many calculations in his temple before the battle is fought. The general who loses makes but few calculations beforehand."

Scott Andrews is a private trader and founder of You can reach him at

comments powered by Disqus