prey were smaller but faster, the amount of energy gained from consuming it
wouldn’t make up for the amount of energy wasted hunting it down. The truth is
that if the animals that make up the better part of a cheetah’s diet disappeared, so
would the cheetah. The same will happen if the environment changes and becomes
more rocky, bumpy, or hilly, or if the forest takes over the savannah, so that the
cheetah can’t utilize its number-one hunting weapon—its speed. Thus, the cheetah
is too dependent on its environment to survive in the long run.
Similarly, a trading system with the characteristics of a cheetah would cease
to work properly if and when the environment it operates in changes ever so slight-
ly (and in a myriad of ways), and the only way for you to find that out is when
you’re already in a drawdown you can’t get out of. Therefore, it is paramount that
your systems can work in as many market environments as possible, or to use the
words of the analogy, find and catch its prey wherever possible, wasting as little
energy as necessary.
Now, this doesn’t have to mean that each system needs to work equally as
well in all conditions. It means that a system needs to work well enough to keep
you afloat, or at least out of disaster in less favorable conditions, while waiting
for the good times to reappear. And the only way to achieve that is to familiar-
ize the system with the less favorable market environments during the research
and building process. More specifically, it means that you should let the sys-
tem’s final parameter settings be influenced by the best settings for less favor-
able conditions, no matter if those less favorable conditions actually produce a
profit or not.
For example: Picture a moving average cross-over system, with a four-day
short average and a 25-day long average, that produces great results in one market
(or market condition), but terrible results in another. For a second market (condi-
tion), the best, but still negative, results might come from a 12-day short average
and an 18-day long average. In that case, you might be better off, in the long run
and on average, with a final setting of nine days for the short average and 20 days
for the long average, which still might produce good results in the first market and
a slight loss in the second market.
Whether you then decide to trade both markets or only the profitable one
depends on other considerations and what you’re trying to achieve with this par-
ticular system in the first place. Assuming you’re only trading the profitable mar-
ket, at least results won’t be as bad as they could have been when the profitable
market starts to behave as the unprofitable one, which you won’t discover until it’s
too late.
If, on the other hand, you trade both markets, you might lower your profits
initially because of the bad performance of the second market. But just as the first
market might start to behave as the second market, the opposite is true as well, and
when that happens, profits will increase. Trading more than one market most like-
ly also will lower the fluctuations of the results and make the system less risky toPART 1 How to Evaluate a System 3