Here’s a note from a fellow reader replying to my post on backtesting without stops. (I always get permission before posting.)
Dave – just to follow up on your backtest comment, I have done some tweaking of my crossing above resistance alert and generated the following results.
Crossing Above Resistance Scan, 2 hour timed exit
Stop Loss Percentage/ Return Percentage
————————————————————–
None/ 9%
3% / -0.3%
5% / 3.5%
8% / 7%
So this scan proves your point that the lower stop loss hurts the result.
Regards,
Tim Welch
Notice the returns for this strategy and how they vary with different stop levels. No stop loss at all had the highest return. The tighter the stop loss the worse the return was.
A well-designed trading strategy will typically exhibit this behavior.
Does this mean you should trade without a stop loss at all?
Wouldn’t that be optimal?
Technically yes that might be theoretically optimal in this case but I never recommend trading a strategy that way.
Why? In systematic trading, you often come across theoretical numbers that suggest a mathematically perfect approach to trading a strategy.
But (and this is a big but), there’s a difference between theoretically optimal and what normal humans can realistically tolerate.
A stop loss gives you an anchor upon which you determine your position size, so you can control your risk appropriately.
It’s also important to remember that a backtest is a simulation over a finite window of market behavior – not a complete picture of everything that can and will eventually occur in the market.
It turns out that predicting the future is hard. (Who knew?)
Using a stop loss is a safeguard against future market behavior that hasn’t been exhibited yet.
For a good example of a situation that would have been impossible to anticipate ahead of time, look at the market action of ZJYL on 12/18/2023.
So how do you choose what stop loss to use for a strategy knowing that whichever one you choose is going to make it worse?
Dealing with these tradeoffs is hard, but it’s a great exercise to examine the worse trades in your backtest and visualize what you would have experienced trading it with a certain size and stop loss.
Now imagine you have traded the strategy for a while, gained confidence in it, increased your position size significantly, and then you have a worst-case scenario. Then imagine you have a few of those bad trades in a row.
It’s not fun to think about, but it’s important to anticipate how you’ll feel and react to this type of scenario.
This will help you visualize what stop loss will make sense to use given the imperfection and tradeoffs.
