....before he puts on a trade?
How much?
Matter of fact, i believe (belief #1 for me), that the answer to this question is the single most decisive factor in a traders equity curve. This is vis a vis the traders bankroll obviously. And this is also obvious to many, already, obviously, lol. But bear with me.
Dr. Van Tharp describes MM as How much? or position sizing. I agree. Ralph Vince gave an experiment to 50 Phd's (barring those with a Trading background or Statistics background), a system that returned 2:1 on a 50% win ration. Awesome, 48 of them went broke or lost money. Aside from the guys who went broke, all scores were different. Other such experiments have been conducted (Van Tharp). What does this tell us?
I already posted on position size vs time frame (http://www.forexfactory.com/showthread.php?t=241268), i was thinking if they were other ways to look at this enticing possibility of variable position sizing (with a baseline so to speak that goes up, hopefully with expectancy). Why am i so concerned with this topic leads really to what my objectives are, quickly i will explain.
I look at trading as a venue that would give me 6% a month on average, i am ready to compound without withdrawal for the next few years. I know this is extemely hard, but that is what my world would look like ideally. I think 3% is more within my reach, but..
see, my personal pain threshold is 500 bets, that means i am comfortable risking 0.2% of my equity on each trade. I know a 100 bets is enough for most, for some even 50 or 20. But for me 500 works. (believe me this a problem really also as quite a few markets like the S&P 500 for example are out of my reach, and i would like to trade a few of these markets to get my trade frequency up)(bankroll size).
So our 1R = 0.2%, and i need to close 30R by the end of the month to hit 6%. Yikes! How am I going to do that? over say 8 years every Month. ( i know i know 200 times ur $, everybody would be a gazziolionaire and all that, but that is not the focus, the focus is longevity, truth in expectations using systems that stand the test of time and conditions(robust).The focus is 6%/3%, lets start assuming for a few min).
So to produce 30R i have to put a minimum of a 100 trades a month, if not 150. (This is highly subjective to each person, my style or whatever u want to call it presents this frame for me, adjust accordingly). Thats like 7 trades a day, that for me IS the problem. I would like to do this in maybe 50 trades a month, if not 40, ha ha the eternal timeframe vs system question (which tf is best?).
So like timeframes, how about systems and correlation and what not. The goal being to up the ante to 1% maybe or 0.5%, with some reasonable level of confidence (quantative also maybe?).
Say for example, I use the following five systems
System A --- DIBS system (daily/weekly and hourly/daily)
System B ---- Pinbar system from J16 (J16/Jankone Daily/5min)
System C --- Dual TF momentum/Elliot/Fib ret Exret/Time Ex & cycle (Robert Miner Daily/1hour)
System D --- BRV Support and Resistance
System E ---- A system constructed around Vantage Point Forex signals
Lets say you decide to bet 0.2% every trade. But I have noticed there are times when more than one system can give me the same signal.
Let me digress here to explain, i know there are a lot of complications here, but lets assume the necessarry precautions have been taken as regards to correlation and coinciding, similar indicators on multiple systems (this is harder then it looks and might be the reason this idea cannot be implemented in the first place, imo, but lets assume for now).
So if i have two systems on the same signal, in the same time whereabouts (scale in or pyramid accordingly on entries that are apart in time and price, or a universal entry trigger gets u in on both at the same time, whatever u like), i go for 0.4%. 3 signals = 0.6% and and so on till 1%.
If i can increase my position size variably like this, i might need 15R to get 6% (bring my avg position size to 0.5%, with the psychological and quantitative (ill get back to this) confidence to do so, hence the edge on yourself). So in effect im looking for 3R from each system as a whole, increasing my position size using, lets call it, global system confluence, helps me!?
Regarding the quantitative aspect here, correlation aside, if two positive expectancy events (assuming i am successful in trading these systems) (lets also assume they have a hit rate of more than 55%) happen at the same time (the signals are the events and they are expected to hit 63% and 60% of the time), don't we improve the overall odds? Am i thinking about all this in a warped way., lol.
I know i am trying to get around the dependency or independency issue of a group of trades, instead trying to fiddle around with my optimal f, but If position size is the most imp Q (how many contracts to put on?)? Then I think getting to know how people think on these lines (either this is explorable or is junk) could be valuable.
All comments appreciated.
How much?
Matter of fact, i believe (belief #1 for me), that the answer to this question is the single most decisive factor in a traders equity curve. This is vis a vis the traders bankroll obviously. And this is also obvious to many, already, obviously, lol. But bear with me.
Dr. Van Tharp describes MM as How much? or position sizing. I agree. Ralph Vince gave an experiment to 50 Phd's (barring those with a Trading background or Statistics background), a system that returned 2:1 on a 50% win ration. Awesome, 48 of them went broke or lost money. Aside from the guys who went broke, all scores were different. Other such experiments have been conducted (Van Tharp). What does this tell us?
I already posted on position size vs time frame (http://www.forexfactory.com/showthread.php?t=241268), i was thinking if they were other ways to look at this enticing possibility of variable position sizing (with a baseline so to speak that goes up, hopefully with expectancy). Why am i so concerned with this topic leads really to what my objectives are, quickly i will explain.
I look at trading as a venue that would give me 6% a month on average, i am ready to compound without withdrawal for the next few years. I know this is extemely hard, but that is what my world would look like ideally. I think 3% is more within my reach, but..
see, my personal pain threshold is 500 bets, that means i am comfortable risking 0.2% of my equity on each trade. I know a 100 bets is enough for most, for some even 50 or 20. But for me 500 works. (believe me this a problem really also as quite a few markets like the S&P 500 for example are out of my reach, and i would like to trade a few of these markets to get my trade frequency up)(bankroll size).
So our 1R = 0.2%, and i need to close 30R by the end of the month to hit 6%. Yikes! How am I going to do that? over say 8 years every Month. ( i know i know 200 times ur $, everybody would be a gazziolionaire and all that, but that is not the focus, the focus is longevity, truth in expectations using systems that stand the test of time and conditions(robust).The focus is 6%/3%, lets start assuming for a few min).
So to produce 30R i have to put a minimum of a 100 trades a month, if not 150. (This is highly subjective to each person, my style or whatever u want to call it presents this frame for me, adjust accordingly). Thats like 7 trades a day, that for me IS the problem. I would like to do this in maybe 50 trades a month, if not 40, ha ha the eternal timeframe vs system question (which tf is best?).
So like timeframes, how about systems and correlation and what not. The goal being to up the ante to 1% maybe or 0.5%, with some reasonable level of confidence (quantative also maybe?).
Say for example, I use the following five systems
System A --- DIBS system (daily/weekly and hourly/daily)
System B ---- Pinbar system from J16 (J16/Jankone Daily/5min)
System C --- Dual TF momentum/Elliot/Fib ret Exret/Time Ex & cycle (Robert Miner Daily/1hour)
System D --- BRV Support and Resistance
System E ---- A system constructed around Vantage Point Forex signals
Lets say you decide to bet 0.2% every trade. But I have noticed there are times when more than one system can give me the same signal.
Let me digress here to explain, i know there are a lot of complications here, but lets assume the necessarry precautions have been taken as regards to correlation and coinciding, similar indicators on multiple systems (this is harder then it looks and might be the reason this idea cannot be implemented in the first place, imo, but lets assume for now).
So if i have two systems on the same signal, in the same time whereabouts (scale in or pyramid accordingly on entries that are apart in time and price, or a universal entry trigger gets u in on both at the same time, whatever u like), i go for 0.4%. 3 signals = 0.6% and and so on till 1%.
If i can increase my position size variably like this, i might need 15R to get 6% (bring my avg position size to 0.5%, with the psychological and quantitative (ill get back to this) confidence to do so, hence the edge on yourself). So in effect im looking for 3R from each system as a whole, increasing my position size using, lets call it, global system confluence, helps me!?
Regarding the quantitative aspect here, correlation aside, if two positive expectancy events (assuming i am successful in trading these systems) (lets also assume they have a hit rate of more than 55%) happen at the same time (the signals are the events and they are expected to hit 63% and 60% of the time), don't we improve the overall odds? Am i thinking about all this in a warped way., lol.
I know i am trying to get around the dependency or independency issue of a group of trades, instead trying to fiddle around with my optimal f, but If position size is the most imp Q (how many contracts to put on?)? Then I think getting to know how people think on these lines (either this is explorable or is junk) could be valuable.
All comments appreciated.