Way of the turtle

154
•
Way of the Turtle

---

in a simulation for a breakout-based system that looked at this trade
there would not have been a breakout and thus no trade.
Now suppose that in the same circumstances you enter the mar-
ket and buy up those 1,000 contracts at an average price of $409.00;
there are no more sellers at that price, and so you have to buy
another 100 from the sellers at $411.00. This trade causes the large
buyer to begin buying, at which time you sell him the 1,000 con-
tracts at $411.00. Although he thinks he got a good price, you made
an excellent trade. All that remains is to get rid of the extra 100 con-
tracts. Since there are no buyers at the recent prices, you have to
sell lower, and so you sell your 100 contracts back where the mar-
ket had been trading: at $407.00. You lose $4 

 100 ounces on
100 contracts, or $40,000, but you made $2 

 100 ounces on 1,000
contracts for a new profit of $160,000 not counting commissions.
Not bad for a few seconds’ work.
What happened to those traders from ACME who had been
counting on the edge in the breakout? They are sitting on a large
position that was entered for reasons different from those their back-
test would have indicated. This is the result of a trader effect.
One specific example of this is provided by a system that became
very popular a few years back because it had maintained excellent
performance for many years. For that reason, lots of brokers started
to offer it to their customers. At one time I heard estimates that sev-
eral hundred million dollars in aggregate was being traded using
this system. Shortly after the system reached its peak in popularity,
those trading it experienced a prolonged drawdown that was much
longer and higher than anything that had occurred in 20 years of
backtesting. The system had an exploitable flaw. If the closing price
passed a certain level, there would be buy or sell orders on the fol-
Lies, Damn Lies, and Backtests
•
155

---

lowing morning’s open. Since other traders knew the price levels
that would trigger those orders, it was a simple matter to buy on the
close ahead of the next morning’s open. One then could exit the
position by selling just after the following day’s open, which would
generally be much higher because of all the buy orders that had
been generated overnight as a result of the system’s rules. 
To make matters worse, the system’s authors used portfolios that
included illiquid markets such as lumber and propane, which
could move quite a bit on relatively light volume, and so many peo-
ple who traded with the system also traded those illiquid markets.
I’m sure that one of the reasons for that system’s sudden unprece-
dented drawdown was exactly this sort of anticipatory buying, which
effectively ruined its edge for a time. Other traders are not that
dense. They will exploit any repeated patterns that they notice. This
is one of the reasons why it is better to develop your own system;
you can build a system with which it is much less likely that you
will have your edge ruined by trader effects because other traders
will not know exactly when you will be buying or selling.
When we traded for Rich, there were often times when we would
all be entering trades at essentially the same time. Market traders
knew that when they started to get large orders from us, the orders
probably would continue for quite some time. For that reason, at
times the floor traders and brokers would start to move the market
ahead of our orders. Since we used limit orders, this was a bit more
risky—that was one of the reasons we used limit orders—because
we would not get filled in those circumstances and so we could pull
our orders. Sometimes when I wanted to buy and knew that the
market was particularly prone to having the locals move it in antic-
ipation of our orders, I would send fake orders in the opposite direc-