better off you are. (Recall the example in the section “Net Profit,” which produced
a higher return than a buy-and-hold strategy, even though it spent less time in the
market and with a lower average profit per share traded.) To calculate the relative
time spent in the market, multiply the number of winning trades by the average
number of bars for the winners. Add this to the number of losing trades, multiplied
by the average number of bars for the losers. Finally, divide by the total number of
bars examined. Time spent in the market is a double-edged sword, because the less
time spent in the market to reach a certain profit, the longer it also might take to
get out of a drawdown.
Let’s get back to the examples about trading your way out of a drawdown pre-
sented earlier. What if the system in question produces approximately two trades
per month and market, and the average trade length is three days? If you only trade
the system on one market, the number of days in a trade to get you out of a $7,200
drawdown is 45 (15 * 3). But because the system only spends about 30 percent of
its time in a trade (6 / 20), the actual number of trading days to get back into the
black is 150 (45 / 0.3). That is more than six months’ worth of full-time trading to32 PART 1 How to Evaluate a System
FIGURE 2.7
Distinctive distribution of trades from successful trading systems.Occurrences