Strategy Lab: Buy break of 20-day high / Sell loss of 20-day low

With the launch of Stock Alerts it is possible to set up simple trading strategies on groups of your favorite stocks; with alerts set for buying and selling. In this test I will look at the performance of buying a new 20-day high and selling on a new 20-day low.


Test period: A complete bull-bear cycle defined by the S&P (March 20th 2000 to October 8th 2007).

I tested on two groups of stocks:

US stocks (Active Trader): AAPL BA C CAT CSCO DIS GM HPQ IBM INTC IP JPM KO MSFT SBUX T WMT

European ADRs: ALU BHP BP SAP DT ASML STM BCS UN TOT ELN AZN DEO RYAAY LUX

Invested: $5,000 per trade

Commission: $9.95

Trades: Round-trip only; partial trades were excluded.

There was a considerable difference in the performance of the strategy depending on whether a protective stop was used or not. In situations where a stop was used there were between 519 and 974 trades compared to the 'pure trading' system, where buy and sells were dependent on the break of 20-day highs and lows, which generated only 379 trades in a seven year period. Unfortunately, the strategy when used with a stop failed to generate a profit when commissions were a factor in the loss calculation; only by entering and exiting trades on the 20-day high/lows was the system profitable.

If a user was to use this strategy with a zero-cost broker there was a significant improvement.

Our European ADRs returned a total profit of $34,847 over the 7 years (Profit factor of 1.45) with the Active Trader a more modest $5,268 (Profit factor of 1.06). Although of the European ADRs, Elan Corp (ELN) and SAP (SAP) contributed $33,270 of those gains.


The success of the strategy was dependent on letting it run without the use of protective stops - a risky play which leaves the strategy prone to slippage (gap breaks in particular).

Test Period: Three random years on Active Trader stocks?

The three time frames selected were April 2007/08 (bearish), April 2001/02 (sideways) and October 2002/03 (bullish).

There was an interesting divergence of performance. When a stop was used the worst returns was given during the sideways market of 2001/02 (-$4,330 to -$6,542) with the no-stop strategy still returning a loss, but below that of a stop strategy at -$2,084. During the bear market (2007/08) the use of stops curtailed much of the losses (-$442 to -$2,373), but the no-stop strategy was hammered at -$6,311. In the bull market only the no-stop strategy was profitable with a return of +$3,336, with losses from using a stop of between -$125 and -$1,132. The huge variation impacted the overall returns to the extent none were profitable, with the win percentage hovering around the 32% mark:


In summary, while not the best strategy it may have merit when used in conjunction with a volume alert to filter weak signals. This is something for a later test.

This article featured in this week's Festival of Stocks hosted by Under the Buttonwood Tree.

Dr. Declan Fallon, Senior Market Technician, Zignals.com the free stock alerts, market alerts, and stock charts website