FW: counterargument to c.lebeau's constant bet size under drawdown

["Brad S. Yoneoka" <byoneoka@xxxxxxxxxxxxx>]

there are several weaknesses in the argument to maintain the same bet size
under drawdown.

(1)  first, is the assumption that the future of a given market will be no
worse than the past and therefore the future success of a trading system
will be no worse than in the past.  particularly under mechanical system
testing, there is an assumption under system testing over whatever
historical period tested that we have seen the worst sequence of losses in
the past.  there is no inherent reason why a trader/investor should accept
this hypothesis.  for ex, in a growth market with increasing volatility as
is the djia and s&ps, there is good reason to expect that the probability of
a sequence of losses in the future will exceed that of the worst in the past
increases.  judgmental assessments that arbitrarily adjusts worst case {like
larry williams' doubling of max drawdown} helps to plan better for possible
future losses, but still is no guarantee that drawdown won't be even worse.

(2)  second, is the problem of a market meltdown.  much of the recent work
on risk measurement over the last 13 years or so of hi market volatility and
market crises (like the 87 stock market crash) have clarified that the level
of risks are significantly higher than popular models estimate. for example,
clearinghouse bankruptcy post-oct 19, 87 was far in excess {est of 50%
higher) of that calculated by the dominant risk model used by the exchanges
to set margins, based the black-scholes option pricing model.  this is no
less true of trading models.

(3)  third, is the implicit assumption of invariant investor/trader utility
curves over account size.  attempts to maximize returns of trading a system
is an assertion of a goal, not a given.  risk/return is an individual matter
and can very greatly from one investor/trader to another.  the experienced
trader knows that the real name of the game is managing risk and minimizing
losses.  in a drawdown, as account size decreases, constant bet size would
increase risk.  and if case 1 or 2 occurs, losses will be greater.  the risk
averse trader, in a drawdown, would decrease bet size.  this would be an
acceptable and rational trade-off of accepting lower future returns by
taking less (or no greater) risk now.

so these considerations: we can't/shouldn't try to forecast the market
{implicit in choosing to trade a given system}, we never know when a market
meltdown will occur, and we can choose the priority of trading to manage
risk--all argue against constant bet size under drawdown and for asymmetric
money management.

regards,
brad yoneoka