Backtesting - fixed lot or fixed risk?
Hi
For real trading the answer is obvious - fixed risk per trade, but whatwith backtesting? Wont it distort the results of EA? For the purpose of ratingthe EA which method is better?
Actually, there are more ways of doing thinks besides fixed lots or fixed risk - there is also the popular "Fractional Risk" (unless that is what you meant), but there are others too.
When back-testing (and forward testing), you cannot just rely on one metric (namely profit). You have to analyse many metrics based on both "Money" profit as well "Pips" profit (which would be equivalent to fixed lots). Looking at Money or Pips alone is not sufficient and neither is "Best" alone, because it depends on the type of strategy. Always look at both in conjunctions with the other metrics, such as Profit Factor, Draw-down, Recovery, the shape of Equity Curve and many others, but always looking at both Money and Pips.
Hi
For real trading the answer is obvious - fixed risk per trade, but whatwith backtesting? Wont it distort the results of EA? For the purpose of ratingthe EA which method is better?
G'day bro. I tend to start my first runs using a balance of $50000 but at 0.1% risk per 100 pip stop loss. ie if I was testing with a 20 pip stop I would risk 0.02%. 200 pip stop 0.2% risk. This way it's hard for any bot to do any real damage to the balance and distorting the results as you perceive.
During an optimization I use always fixed lots. This guarantees that the 'money-results' (which is essential for the optimization) of the first and the last trade are comparable!
If I would plan to sell an EA I would use a fixed risk!
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
For real trading the answer is obvious - fixed risk per trade, but what with backtesting? Wont it distort the results of EA? For the purpose of rating the EA which method is better?