I have manually traded a system on price data from January 2010 to December 2010.
I have well over 100 trades.
It is profitable but I do not only want to optimize it, I have to, because it did not do what it was designed to in the second half of the year.
So I have to find out why and if I can tweak it.
I have made the test with ForexTester and have no other trading software. I particularly don`t have any backtesting software that can try different variations for several parameters. It could become extremely time consuming to manually do this but I am willing to do so. Although it wouldn`t be smart not to at least ask if there are more time efficient ways to do this.
I have LibreOffice`s/OpenOffice`s spreadsheet application at my disposal.
What I was going to do now was is going through my list of questions for all trades and answer these questions:
1.
How close did the price got to my stop loss in the winning trades?
Do the bigger winning trades differ from the smaller winning trades in this aspect?
Can I use a smaller SL to increase profits particularly on the big winning trades while not losing much on those trades that would now get stopped out? Or are more smaller trades more profitable for this system within the tested time period?
2.
What is the percentage of winning trades vs. losing trades and profits vs. losses at different times/weekdays?
3.
More or less profits when using only or additional trailing stop instead of exit at MACD-cross? How close should the trailing stop be?
4.
I always open 2 positions for each trade. One is supposed to regularly capture small wins the other the occasional big trends.
More profits if doubling the position once first target is hit?
5.
The system used EMA-crosses and MACD-crosses as signals for entry and exit. Here the question would be by how much a cross qualifies as signal and if a failed downward cross for example would be a good upward signal.
The parameter settings for EMAs and MACD would probably mean an incredible amount of work. For the start I don`t intend looking into these although I have the idea that it might be worth it.
6.
Optimal TP for position 1?
7.
More profit or less if SL of position 2 is set to break even once TP of position 1 is reached? This means I am at that point trading with no risk.
Sounds good but might be worth exploring if opening another position at that point in time would create more profits or more losses.
I now want to
(a)
create a spreadsheet with the parameters I need to answer these questions
and then
(b)
enter all these parameters for all trades.
Later on I will add calculations to the spreadsheet document to produce some results for another test run.
If I change parameters I guess I will have to calculate all trades again. Currently I don`t see any other way. Again, I am willing to do this but any hints that save time are appreciated a lot.
If I have forgotten anything or if you think my approach is inefficient to start with please let me know.
I have well over 100 trades.
It is profitable but I do not only want to optimize it, I have to, because it did not do what it was designed to in the second half of the year.
So I have to find out why and if I can tweak it.
I have made the test with ForexTester and have no other trading software. I particularly don`t have any backtesting software that can try different variations for several parameters. It could become extremely time consuming to manually do this but I am willing to do so. Although it wouldn`t be smart not to at least ask if there are more time efficient ways to do this.
I have LibreOffice`s/OpenOffice`s spreadsheet application at my disposal.
What I was going to do now was is going through my list of questions for all trades and answer these questions:
1.
How close did the price got to my stop loss in the winning trades?
Do the bigger winning trades differ from the smaller winning trades in this aspect?
Can I use a smaller SL to increase profits particularly on the big winning trades while not losing much on those trades that would now get stopped out? Or are more smaller trades more profitable for this system within the tested time period?
2.
What is the percentage of winning trades vs. losing trades and profits vs. losses at different times/weekdays?
3.
More or less profits when using only or additional trailing stop instead of exit at MACD-cross? How close should the trailing stop be?
4.
I always open 2 positions for each trade. One is supposed to regularly capture small wins the other the occasional big trends.
More profits if doubling the position once first target is hit?
5.
The system used EMA-crosses and MACD-crosses as signals for entry and exit. Here the question would be by how much a cross qualifies as signal and if a failed downward cross for example would be a good upward signal.
The parameter settings for EMAs and MACD would probably mean an incredible amount of work. For the start I don`t intend looking into these although I have the idea that it might be worth it.
6.
Optimal TP for position 1?
7.
More profit or less if SL of position 2 is set to break even once TP of position 1 is reached? This means I am at that point trading with no risk.
Sounds good but might be worth exploring if opening another position at that point in time would create more profits or more losses.
I now want to
(a)
create a spreadsheet with the parameters I need to answer these questions
and then
(b)
enter all these parameters for all trades.
Later on I will add calculations to the spreadsheet document to produce some results for another test run.
If I change parameters I guess I will have to calculate all trades again. Currently I don`t see any other way. Again, I am willing to do this but any hints that save time are appreciated a lot.
If I have forgotten anything or if you think my approach is inefficient to start with please let me know.