Calculating probabilities such as the one posed in this question is a futile effort when it comes to trading if you ask me. Maybe, over the period you have tested your system it did provide you with an 80% win ratio and whatever r:r. This does not mean however that this will hold true for the future. There are just too many variables to consider when it comes to the markets.
Lets consider trying to calculate these type of probabilities for the flipping of a coin. We can make sure the coin is weighted properly, it is flipped in the exact same manner, wind, temperature and any other variable that may affect the outcome can be controlled. Therefore we can calculate the probabilites of a series of flips and expect them to hold true in the long run. There is the chance of statistical anomalies due to random outcomes, but all this can be calculated and no external forces or outside variables can change these probabilities. It is all random.
This is not true for a trading strategy however as we can't control all the variable that may effect the outcome. There are psychological factors within yourself, and then there is the landscape of the market at the time. Some strategies work well in trending, some in ranging, some in volatile and some in stable markets. The problem is the market is always changing and thats why I think that trying to calculate these statistical probabilities for trading strategies is very misleading. An 80% winning strategy (based on past results) could easily have 10, 20, 30+ losing trades in a row even though the probabilities say this is extremely rare. Some may say that this is due to a statistical anomoly, and maybe they are right this time. But it is far more likely that the market landscape has shifted or, if trading a discretionary strategy, you are 'off your game'.
So to be able to calculate reliable probabilities for a trading strategy one would have to include all the posible variables in the math formula, of which there are just to many.
Lets consider trying to calculate these type of probabilities for the flipping of a coin. We can make sure the coin is weighted properly, it is flipped in the exact same manner, wind, temperature and any other variable that may affect the outcome can be controlled. Therefore we can calculate the probabilites of a series of flips and expect them to hold true in the long run. There is the chance of statistical anomalies due to random outcomes, but all this can be calculated and no external forces or outside variables can change these probabilities. It is all random.
This is not true for a trading strategy however as we can't control all the variable that may effect the outcome. There are psychological factors within yourself, and then there is the landscape of the market at the time. Some strategies work well in trending, some in ranging, some in volatile and some in stable markets. The problem is the market is always changing and thats why I think that trying to calculate these statistical probabilities for trading strategies is very misleading. An 80% winning strategy (based on past results) could easily have 10, 20, 30+ losing trades in a row even though the probabilities say this is extremely rare. Some may say that this is due to a statistical anomoly, and maybe they are right this time. But it is far more likely that the market landscape has shifted or, if trading a discretionary strategy, you are 'off your game'.
So to be able to calculate reliable probabilities for a trading strategy one would have to include all the posible variables in the math formula, of which there are just to many.
