Strategies 5 and 6 used the trailing-stop version of the Harris 3L-R pattern
variation and the stop-loss version of the expert exits systems. The results from
these two strategies were not satisfactory, probably because the systems turned out
to be too short term for our purposes. When a system is “too” short-term, with a
low risk–reward relationship, the profits won’t make up for the costs of trading
with slippage and commission included in the calculations. In this case, we prob-
ably also need to rebalance the distance between the entry price and the money
management point, and the distance between the entry price and the stop loss, to
optimize the number of positions we can be in simultaneously.
Strategy 7 produced satisfactory results, although the current bear market
once again forced the strategy into a new maximum drawdown. For Strategy 8, the
results could not be considered satisfactory. It is apparent that the shorter-term
systems—the Harris 3L-R pattern variation and the expert exits system—do not
work as intended.
Strategy 9 produced satisfactory results, despite a rather steep recent draw-
down. Substituting the 10 Dow stocks in Strategy 9 with the same short-term sys-
tems as in Strategy 8 resulted in very choppy results for Strategy 10. Although the
profit potential is there, it is doubtful a strategy like this should be trusted for real-
time trading—at least not when traded on only a handful of markets.
The last two strategies used a combination of the relative-strength bands and
the rotation filters. For Strategy 11, this resulted in a smooth and steady initial equi-
ty growth that again was weighted down by a larger than tolerable drawdown over
the last few years. For Strategy 12, the results were too choppy and the profits too
dependent on the bull-market run of the late 1990s. As was the case for Strategy 10,
it is doubtful a strategy like this should be trusted for real-time trading.
Looking through the overall results, three things come to mind. Number one
is that the shorter-term systems, the Harris 3L-R pattern variation and the expert
exits, did not work as intended, probably for two reasons: Reason one is that their
short trading horizons limit the profit potential. Reason two is that the money
management point probably is too close to the entry price, given the average stop-
loss distance, to allow us to be in an optimal number of positions at one time.
Reason two can be dealt with simply by shifting the MMP further away from the
entry price; reason one probably calls for a revision of the research surrounding
the stops and exits.
The second thing that comes to mind is that the relative-strength bands and
the rotation filters allowed us to test only 20 stocks, which limited our diversifi-
cation possibilities. It is likely the risk-adjusted returns would have been higher for
Strategies 7 to 12 had we been able to test them on more stocks. Talking about the
filters, it’s also worth mentioning that several of the systems might have func-
tioned better without them. I leave that for you to decide.
Last, but not least, most strategies probably should have done much better
trading the short side as well, despite the fact that it would have resulted in a slightCHAPTER 29 Consistent Strategies 383