Monday, April 27, 2009

Regaining Trading Consistency

from Dr. Brett--
A reader recently wrote to me the following:

I was a successful consistent trader who always hit singles and doubles ($1000-$3000 a day) for 48 months in a row without having a losing month (1999-2003).Then one day I struck out. I lost $38,000 in one stock and had my first losing month as a trader ever. Since then I have not had two consecutive winning months and in fact have only had a handful of profitable months since then. I am still looking for the road back to consistency. No matter how close I get I always find a way to screw it up even if it is on the last day of the month. Or I give back the month with just some silly unimportant trade that turns into a disaster. It is like I subconsciously look for these situations just so I can mess up.

This is not such an unusual scenario. One large loss can trigger a cascade of attempts to make back the money, further mistakes, and expanding losses. The key is breaking this cycle of losing money, attempting to make the money back with aggressive trades, and continuing to lose.

The first thing I'd have our trader look at is where he is placing stops and targets for his trades. Note that his successful period was 1999-2003. That was a period of much higher price volatility than we've seen since then. What constitutes "singles and doubles" in a high volatility environment is a home run trade in a slow, low-volatility market. It is entirely conceivable that our trader is placing targets too far from his entries, allowing small gains to reverse on him. Similarly, he may be letting trades get too far away from him simply because he is calibrated to a higher level of volatility.

A good way to test these hypotheses would be to study trades over the last several months. If losing trades are larger than winners on average, and if many losers start out as winners, that would suggest that our trader needs to adjust to the post 2003 environment.

To break the cycle mentioned above, the first step is to drastically reduce trading size. I would cut size to 1/4 the average at the most. The goal is to keep a little skin in the game, but take P/L (and the push to make back money) off the table temporarily. The initial objective is not to make money, but to regain a trading rhythm by getting back to singles and doubles.

The next step is to identify those singles and doubles. That means deconstructing the account statement and identifying which trades are making money and which aren't. I would break the data down into time of day, stock/index being traded, long/short, and size. I would also look to see if there are large outlier trades to the downside that are pulling down P/L, and if there are some trades that are making money consistently.

Once our trader has identified what's working, the idea is to keep position size fixed and *only* trade those setups that have been working. This is the foundation to build upon. These setups can be written down and mentally rehearsed ahead of the trading day to build consistency. The idea is to not increase size *and* not trade other patterns until consistency is achieved with smaller size and the most successful setups.

There is only one cure for trauma, and that is repeated experiences of control and safety. We want trading to be routine, not highly emotionally charged.

Finally, I would encourage our trader to take a look at how he is viewing his situation. Note above that he talks of the $38,000 loss and the silly trade that "turns into a disaster" as if these are things happening to him, not things that he is actively doing. A simple strategy would be to have the trader write down the four things he is responsible for prior to each trade:

* The Entry
* The Target(s)
* The Stop
* The Position Size

We can't control whether any individual trade will be a winner, but we can control how much we are willing to bet on each trade. Outsized losses don't happen to a trader; they are actively caused. It is harder to allow those things to occur if you're talking aloud those four trade parameters and have them written in front of you.

So there it is in a nutshell. My advice is to get small, get selective, and take responsibility for what can be controlled.

Do readers have additional advice? Let's see if we can help a reader. Thanks!

Brett