List member Dean J. wrote in with a question on position sizing (used with permission):
Dean J.
When you are doing your backtesting, do you use a standard size for each stock, ie, every trade is 100 shares, or do you size by a % or something else? In other words, a $20 stock won’t move as much as a $100 stock if they were all the same size, wouldn’t it make the results look different than the true value of each stock might be in your strategy?
Dave:
There are a few ways I’ll size positions based on the strategy:
- Constant shares (rare, but there are strategies where this is the best choice)
- Constant dollars – e.g. $1000 / entry_price
- Based on ATR or volatility – e.g. $1000 / ATR
- Based on the stop distance – when signal is from a pattern with a logical stop
I have strategies that use all these approaches.
I think the most underrated is the last one – sizing based on the stop distance.
I see some really good traders who don’t even consider sizing based on stop distance.
The reason it’s so valuable is that the sizing is fundamentally based on the trading signal.
When you do that and then optimize your strategy in the Cruncher, you’ll get a completely different set of predictive variables.
And it’s easier to find ways to scale the strategy – because you’ll have trades where the stop is a short distance away.
And because it’s based on a pattern, the setup itself (and the sizing) leads to a strategy with a clearer path to confidence.
That is, you have a path to believe in it enough to trade with significant size.
-Dave
