Way of the turtle

The
N Factor
As was mentioned earlier, Rich and Bill used an innovative method
for determining the position size for each market on the basis of
the amount that market moved up and down each day in constant
dollar terms. They determined the number of contracts for each
market that would cause them all to move up and down approxi-
Risk and Money Management
•
117

---

mately the same amount in dollar terms. Since the number of con-
tracts we traded for each market was adjusted for this volatility
measure, N, the daily fluctuations for any particular trade tended
to be similar.
Some traders prefer to measure risk in terms of the distance
between the price at which one will exit a trade and the price at
which that trade was entered. That is only one way of considering
risk. In October 1987, it did not matter where our stops had been.
The market gapped through our stops overnight.
If I had been using a method that relied only on the distance
between entry and stops, I would have lost four times as much as
the typical Turtle on that day because I used a stop that was one-
quarter the size. I used a 
1
⁄
2
-ATR stop, whereas most Turtles used a
2-ATR stop. Thus, if I had been using a method that sized purely
on the distance to the stop, my calculations would have resulted in
a position that was four times larger than that of the typical Turtle.
Fortunately, Rich used volatility-based position sizing as a way
to manage risk, and so I had the same position sizes relative to my
account as did the other Turtles, and our exposure to the price
shock was the same. I am certain that this method was not acci-
dental. Surely Rich and Bill both recalled experiencing prior price
shocks when they determined how to limit the Turtles’ maximum
allowable risk levels.
One of the smartest things Rich and Bill did when they gave us
our trading rules was to impose overall risk limits on us. This had
important implications for our drawdowns and particularly for our
exposure to price shocks. As was mentioned earlier, we put our
positions in chunks we called units. Each unit was sized by deter-
mining the number of contracts where 1-ATR of price movement