I was going to start a series on the markets tonight. But, Peter Whitnall’s comment on the Ebb & Flow raised quite a few e-mails.
In this blog, I’ll answer the question: how do we know when to increase and decrease our position size?
Well first off, we can never ‘know’ but we can make solid guesses. The guesses will always be late. It’s a bit like a moving average, trend identification, system: we’ll catch the turns a little after the event. Secondly, the approach I use depends on keeping detailed records of my trades – details that allow me to calculate the Expectancy Return and Expectancy Ratio from the results.
I’ll illustrate the ideas using the Expectancy Ratio – the same process applies to the Expectancy Return.
Figure 1 below shows the statistics from my Expectancy Ratio returns. They show that 68% of the time my Reward:Risk Ratio will range from 0.80 to 2.34. This is a key statistic. I calculate the Expectancy Return after each completed trade based:
- from historical record and
- from the beginning of each month.
The historical record is my standard; the monthly results are the tracking device.
As long as the closed out trades show a Reward to Risk with 0.80 to 2.34, then I treat the market as being in its Normal State – some wins, some losses. If I see a number of trades below 0.80 to .41, then I reduce size. This area represents a withdrawal of the market from my plan and is an amber light. If I see a number of trades below .36, I stop trading or reduce size to the smallest position I am willing to take. This area is where the market has totally withdrawn from my plan and it represents a red light – the full Ebb State.
On the flip side, when I see a series of trades at or above 2.4 and below 3, I become more aggressive in my position sizing: I’ll take up to 1.5 times my normal size. If I have a series of results above 3.60, I’ll go up to 2.5 times normal size.
No magic – just keep the stats and watch the Expectancy Return and Expectancy Ratio.
- Expectancy Return: (Avg$Win x WinRate) – (Avg$Loss x LossRate)
- Expectancy Ratio: (Avg$Win x WinRate) /(Avg$Loss x LossRate)
FIGURE 1 Expectancy Ratio Stats