relatively slow equity growth in dollar terms might not be enough to increase the
number of shares traded.
However, the way I see it, the critics overlook two important points. First,
nothing forces anyone to trade the exact optimal f. The important thing is to not
take any unnecessary risk by trading above it. Also, never forget that the optimal f
value is based on historical or back-tested trades. As already stated, chances are
Murphy’s Law will make sure that the system will not work as well in the future
as it has in the past—at least not initially—which means that the true ffor that
period of trading will be lower than the historical f. Therefore, it’s better to be safe
than sorry and trade at an fconsiderably lower than the optimal fderived from his-
torical data. To do that, you need to know where fis in the first place.
To trade below the historical optimal f, you need to decide how much below
you must go to achieve a suitable risk–reward relationship. Remember, the less
you risk, the less you will lose and the smaller the drawdowns will be. For exam-
ple, it is fully possible to calculate an fthat aims to find an equity curve that is as
smooth as possible in relation to the equity growth, or one that simply aims to min-
imize drawdowns and/or flat times, or....
Second, in Ralph Vince’s original version of the formula, fdepends on the
worst historical loser. We use the largest historical loser in the formula for the HPR
because we implicitly assume that one reason why we’re trading this system in the
first place is that its largest historical loser is within the limits for what we can tol-
erate given the system’s overall profit potential, but also because the loss has
already happened (whether for real or during testing), which, because of its size,
makes it less likely to happen again.
Note, however, that the “better” the system, the higher the optimal fwill be,
because the better the system, the more we should be able to risk per trade to make
the most out of the system by maximizing its profit-generating potentials.
However, if we ever were to encounter a loss of the same size as the worst histor-
ical loser, so that Profitnequals WCS, then the HPR for that trade also will equal
f, according to the formula HPRn 1 f. Thus, the better the system, or the high-
er its profit-generating potential, and the higher the f, the more you stand to lose
on one single trade, should the system once again experience a loser of the same
magnitude as the worst historical loser.
I know of traders who claim to have systems good enough that they trade
them with an fof 0.20, meaning they’re risking 20 percent of their capital in any
one trade. Is this a wise thing to do, no matter how good the system is? No, of
course not. Risking as much as 20 percent per trade means that it will only take
three losing trades in a row to deplete the trading capital by close to 50 percent.
That is, after three trades, losing 20 percent of your capital in each trade, you’d be
in a 50 percent drawdown. To get back to where you started you would need to
increase your current capital by 100 percent, for which you’d need at least four 20
percent winners in a row—how likely is that?300 PART 4 Money Management