Wednesday, October 28, 2009

The Essence of Trading Psychology

Let's see if I can capture the essence of trading psychology in a single image.

Imagine the distribution of an active trader's daily P/L across a calendar year. There are peaks near the center of the histogram, with many days of relatively small winning and losing days. There are fewer big losing and big winning days.

How fat and extended the tail is at the right side of the distribution tells us how many big winning days the trader has had. This is a measure of aggressiveness and risk-taking: the ability to press an advantage when it's present.

How fat and extended the tail is at the left side of the distribution tells us how many big losing days the trader has had. This is a measure of discipline, prudence, and self-control.

How many winning vs. losing days the trader has is a measure of edge: one's ability to find opportunity in markets across market conditions.

The ability to go for the knockout when you have the market on the ropes; the ability to play defense when the market is fooling you; the ability to stay mentally flexible and find an edge across the many market conditions that occur during a year: all of trading psychology boils down to those virtues--and the resulting shape of one's P/L distribution.


Matt Fahmie said...

Imagine a trader who's central
distribution was winning days...

How would you define that trader's
psychological make up?

Ross said...

If I have imagined Dr. Brett's image correctly then having the highest frequency of trades (the 'central distribution') as winners, without changing the distribution's shape, would shift the plane to the 'left' where smaller losses and larger winners would be less frequent and large losses may not exist.

The low frequency of small losses and large winners would probably be the profile of someone with a low risk tolerence, someone who 'chokes their trades' cutting their winners before they fully run their course and cutting their losses very quickly. This person would probably not go broke trading but it is unlikely that they would make super-normal returns amount either. It would depend on the magnitute (not frequency) of the small wins versus the small losses.

TradeMind said...

Great Post as always!!
You could calculate the skeweness of your P/L distribution, track it over time and use it as a performance metric.