Stoploss & Profit targets for algotrades

[SenorQuant]

Hi everyone,

I have been doing a lot of back-testing and reading books about Algotrading and it seem that the most logical place to put a stoploss would be to consider market volatility at the time and calculate your position size according to the stoploss.

In example:

• Price broke out of 10 day high.
• Risk is 1% of equity on each trade.
• Stoploss should be set for example at 1*ATR of the 10 day EMA (These are just hypothetical examples) and the position size would be the difference between entry and stoploss divided by the Risk amount.

Many books and traders agree on this risk management method however I have found by backtesting that a fixed percentage stoploss with fixed percentage profit target have been delivering equally good and in some cases even better results.

Are fixed percentage methods highly irrational to use cause the don´t consider market volatility and should they be avoided even though results have suggested they do actually work?

Best regards,
SenorQuant

 

[Almarok]

If it is helpful, on this page Kevin Davey (former World Trading Champion), demonstrates with various tests that fixed Stoploss is better than using ATR:

Best Algo Trading Tip: Dollar Stop vs. ATR Stop Loss

Get the latest algorithmic trading tip. We dive deep on ATR Stop Loss and Dollar Stop.

kjtradingsystems.com

 

[tomorton]

First off, I'm not clear what "1*ATR of the 10 day EMA" actually means. ATR is a calculation derived from recurring price ranges, not the MA. MA does not have a range, its a unique value.

However, taking the line you've opened here, a fixed percentage stop-loss distance on a given market might well be totally workable and as profitable as any other technique. Except that how would you know what percentage to use in the first place? Would the right percentage be 2.5% or should it be 25.0%? Or one of the multiple values in between or outside that range?

And if 2.5% is the right distance on EUR/USD, how would that help you decide the right distance on USD/JPY?

And how does a percentage of the distance to the profit target relate to anything in the real world and not some figure dreamt up in the trader's head?

 

[Almarok]

Using Trading Systems one can calculate the best Stoploss and Profit target values based on each individual Market;

so the Stoploss and Profit target on EUR/USD will be different from those on USD/JPY and will be different from those calculated on Tesla or on the Mini S&P, and will have different values when calculated on Futures because of leverage, etc.,...etc.

 

[morehastelessspeed]

I find ATR generally useful as a metric of volatility; ATR is ready understandable and widely available. It helps me stay grounded on some objective measure while trading price action visually, not to dream grossly unrealistic or "irrational" TP or ST distances.

However, to decide on it as an element of a systematic strategy, I would backtest the strategy as a whole. It might work better than fixed percentage in the context of one strategy, but not if used as an element of another strategy, even if the latter strategy might appear similar to the former.

Also, you can combine both, ATR and fixed percentage in your calculations. Keeping in mind that the risk is not just the ST distance, some traders might choose to reduce size when volatility increases because they expect greater slippage.