Now that you have back tested your robot for the last 12 months let’s assume that you get the following statistics.
Bars in test | 11257 | Ticks modelled | 14846077 | Modelling quality | 90.00% |
Mismatched charts errors | 10 | ||||
Initial deposit | 2000.00 | Spread | 50 | ||
Total net profit | 2287.30 | Gross profit | 4554.62 | Gross loss | -2267.32 |
Profit factor | 2.01 | Expected payoff | 25.41 | ||
Absolute drawdown | 81.56 | Maximal drawdown | 763.43 (21.43%) | Relative drawdown | 21.43% (763.43) |
Total trades | 90 | Short positions (won %) | 30 (80.00%) | Long positions (won %) | 60 (81.67%) |
Profit trades (% of total) | 73 (81.11%) | Loss trades (% of total) | 17 (18.89%) | ||
Largest | profit trade | 105.46 | loss trade | -243.52 | |
Average | profit trade | 62.39 | loss trade | -133.37 | |
Maximum | consecutive wins (profit in money) | 20 (1385.47) | consecutive losses (loss in money) | 3 (-106.93) | |
Maximal | consecutive profit (count of wins) | 1385.47 (20) | consecutive loss (count of losses) | -251.12 (2) | |
Average | consecutive wins | 5 | consecutive losses | 1 |
You would expect that the worst case would be to experience
- – 21.43% of your capital any given time
- maximum 3 consecutive losses in a row
- 1 consecutive loss on average = on average you would expect a wining trade after a losing trade
What would you do?
What should be your reaction if you experience 4 consecutive losing trades (one more than the statistics ) in a row and you are already at negative territory at your 5th trade?
A. Would you stay cool? (Keep your stop limit intact and follow the same strategy even if you were consecutively stopped out for the fifth time)
B. Would you close the 5th trade and get rid of your strategy?
Your best option is to consult the statistics.
You need to mathematically determine if the results you get is significantly different from the statistics of your backtest. My service would prove to be valuable at this stage.
Because there are times when number 4 is not significantly different from number 3 and there are times when their difference is significant.