Thanks to all who put pen to paper yesterday. My purpose was to make clear my reasons for writing and I am happy so many share them with me.
In this series, I’d like to talk about position sizing and suggest certain formulas.
Position sizing is part of the Money Management factor in the factors for trading success: Written Plan with an Edge x Effective Money Management x Winning Psychology. The multiplication sign is important. It emphasizes that you need to be competent in all three areas – incompetence in any one area will lead to failure. As an extreme example, if you have ‘0’ competence in any area, then your dollar return must be ‘0’.
I feel that much of what is written about money management misses the point – the authors speak of measuring the outcome in dollar terms. Yet, a dollar outcome fails to account for the differing volatilities which differ in the same instrument over different timeframes, and among different instruments. Hence, I believe we need to have a section in our reports that normalizes results. Now sometimes the position sizing formula itself will do the normalizing e.g. the Turtle formula I’ll speak about that later in the series; but most times, we need to input it somewhere in our spreadsheet.
I know of two ways of normalizing:
- Using the ATR or
- Expressing the outcome of the trade as percentage of the initiating price. We do this on a one contract basis. For example: if I sell three contracts of the ES at 1358 and cover at 1348, I have made 10 points per contract. This is a result of (10/1358)% = 0.74%
Either way does the job.
So step one in a management algorithm is to normalize your output.
Step two is to decide on the amount of risk to take per trade. This is both a subjective and objective exercise. The subjective one is more difficult to assess. One of the more interesting ways of doing this is via a visualization question and answer process:
- Select an amount of money that represents a significant return to you but that is within reach. It could be US$100k or US$1M or US$1b.
- Select an amount of money whose loss would mirror (1) except you’d feel pain rather than pleasure. Express this as a function of your capital.
- Assume you have an indeterminate amount of black and white balls in a jar. If you pull the black ball, you lose (b); if you pull the white, you win (a).
- Starting at a percentage you know will be painful – for me it would be 30 & ask: would I choose the game (where I stand to win $X or lose Y%) or not play. If the latter, choose a lower percentage and keep playing until you chose the game. The percentage before you chose the game is your subjective risk threshold. For example: Say at 3% I chose the game, and the percentage before that was 4%, 4% is my threshold.
This provides the threshold for an individual trade, you now repeat the same exercise for consecutive losses and again for portfolio risk. It’s important to do this because each measures a different impact. For example your pain threshold per trade may be 4% but 20% for consecutive losses; and it may be 12% for portfolio loss. In other words, you may not be happy losing more than 4% on any one trade but it would be OK to lose up to 20% of you lost it because you lost on five trades, each losing 4%.
You may be thinking – that’s illogical! I agree but we are talking here of subjective risk assessment, logic has little to do with it. By identifying and keeping below the subjective risk threshold, we ensure we reduce anxiety and thus ensure we keep to our rules.
For the process to work, you need to vividly imagine the pleasure and pain – in short engage your emotions!
So now you have your subjective risk threshold, part of the process to determine the maximum risk per trade. Tomorrow I’ll continue with this thread and look at the objective risk thresholds.