In the same period that BAH returned 82.81 points, our system returned only 32.34 points of profit. If the time in the trades are examined, BAH was in a position for 1000 periods, but our system was in a position for only 283 periods. If the results are extrapolated 1000 periods of being in a position, then the potential profit would be at over 114 points.

But this can be enhanced by changing the exit strategy from a reversal to a trailing stop method. One of the notables of the Parabolic SAR is that it typically works much better as a Parabolic Stop only instead of the Stop and Reverse concept (remember, the completion of one trend does not constitute the beginning of a new one). If this method is utilized on the current system as a trailing stop, then the system would change to this –

Enter Long: ADX(18)>Ref(ADX(18),-1) AND PDI(18)>MDI(18)

Close Long: Cross(SAR(.02,.2),C)

And the equity curve of the new system (red) compared to the equity curve of BAH (green) would look like this:

In the same period that BAH returned 82.81 points, our new system returned only 39.41 points of profit. If the time in the trades are examined, our system was in a position for only 254 periods. If the results are extrapolated 1000 periods of being in a position, then the potential profit would be at over 155 points compared to only 114 points for the previous version of the system.

At this point, may people are curious as to the profitability of the trailing stop methods as stand-alone systems, so here are the two systems –

High-Low Stop

Enter Long: Cross(H,Ref(HHV(H,5),-1))

Close Long: Cross(Ref(LLV(L,5),-1),L)

And the equity curve of the High-Low Stop system (red) compared to the equity curve of BAH (green) would look like this:

This shows that the total profit for the system is 53.69 points, and the total trading time is 676 periods. Extrapolated to 1000 periods comes to 79.42 points. That is 3.39 less that the BAH method. Using a stop as a reversal in this case showed to be less profitable that just holding on. Interestingly enough, a variation of this trailing stop method became the foundation for Richard Dennis’ famous "Turtles" who were all effectively profitable in the commodity markets (they had different buy and sell time lengths to create a breakout trading system that was very successful, especially for its time).

The basic Parabolic SAR Crossover system that was used as the stop of the second test is below –

Parabolic SAR Crossover

Enter Long: C>SAR(.02,2)

Close Long: C And the equity curve of the Parabolic SAR Stop system (red) compared to the equity curve of BAH (green) would look like this:

This shows that the total profit for the system is 46.34 points, and the total trading time is 621 periods. Extrapolated to 1000 periods comes to 74.62 points. This is even less that the High-Low Stop method. Using a stop as a reversal in this case showed to be just as problematic as in the previous case.

It becomes clear that in many (if not most) cases, creating a specific stop or exit strategy becomes much more profitable per unit of tradable time that a simple reversal of conditions. While this knowledge has been promoted by many, actual examples and scenarios backing up these philosophies have been few and far between. Hopefully, with this type of understanding, you will have an added perspective in system design and testing that may not have been available before.