Broker/Data: Mirus (Broker), Continuum (Data), Dorman (Clearing)
Favorite Futures: Futures
Posts: 202 since Mar 2013
Thanks: 426 given,
Tick Counter; help with improving stops and profit targets
I was curious if someone could chime in on helping me track down a possible solution for counting ticks after trade entry.
I am currently using a X stop and A, B and C profit targets. Upon trade entry, I am interested in automatically counting the number of minus ticks (ie, towards my stop) and positive ticks (ie, towards my profit targets) such that I can then build a running record of the relative frequency that negative prices saw B/E to my stop (ie, 0 to -X and all the numbers in between) and B/E to my various profit targets (ie, 0 to +A, +B and +C and all the numbers in between) before the move ends (ie, either -X or +C is hit). My goal is to then use this information to continually be using the highest probability stops and profit targets relative to what price has historically had a higher propensity to do after my specific trade entry.
So for instance, let's say that I have a -5 stop and +4, +8 and +12 profit targets. Let's say that I've been trading this for a while, but then I go back to my tick log (above), and I find that +4 and +5 both have a 63% chance of being hit, +7 has a 33% probability while +8 has a 27% probability and +12 through +18 have the same hit rate of 4%. In that case, probability says that I might want to consider revising my profit targets from +4, +8 and +12 to +5, +7 and +18. I would also revise the stops in a similar observed probability manner.
The goal here would be to insure that I am continually using the optimum stops and profit targets to create the maximum aggregate weighted average theoretical expectancy.
So the reason that I posted this thread in the NT Programming forum instead of the Money Management forum is that I am using NT, and I am hoping that I can figure out a way to create NT code to accomplish the above. I would be willing to take a stab at the coding, but before I started out I first wanted to ping the NT Programming community to see if this has already been done--or just to get ideas.
Please let me know if you have any questions.
I appreciate your input, and thank you for your help.
I don't know if the following will be helpful but until someone else chimes in maybe worth a look. Am wondering if the Tradervue product I've seen on the site of late will do this. Would be a quick matter to ask the CEO in the AMA thread.
I had similar questions regarding to optimize the strategy respectively to determine the differences between the markets. Because i didn't find a proper solution i developed an own trade analysis software which is closely correlated to my strategy. From within NT only i export the trade entry data and the stop and profit targets and the tick/range bar values. Then i combine them with different trade management/exit logics. This helped me to find the most stable Parameters and to get an intimate feeling about how the strategy works in different markets. Deviating from your pure statistic approach i prefered to let the software found the optimum trade exit targets and stop trailing values with brute force of combining all possible values and ranking the overall profit results and the drawdowns. Then these values are the starting point for deeper analysis.
I don't know if i can help you with my answer, but to spend some serious amount of time to learn C# development to build your own tools can solve many questions which can't be solved within NT easely.
The following user says Thank You to Koepisch for this post: