achieving consistent and slow equity growth with controlled risk?

Discussion in 'Strategy Development' started by jjk2, Mar 13, 2008.

  1. jjk2

    jjk2

    NOTION: by having an unlimited upside potential and limited risk calculated by using our total expected loss in long time frames (2 weeks and up), we can achieve theoretically, SLOW but somewhat consistant equity growth in volatile assets.

    actually, this system i dont think is possible for retail traders due to execution and slippage.

    the actual mechanics is pretty much brain dead. u buy or sell when ur pending orders are hit at the highest high and lowest low during london session.

    its not unusual to observe that on a weekly basis, the GBP pairs makes at least one significant move in one direction in one of the 5 days of the week.

    in 2 weeks, its more probable that a huge move will come.

    in 3 weeks, the probability is even higher.

    as you increase your timeframe, we can observe the probability of a big move coming in a single day is quite large, for obvious reasons, the unexpected outcome is undervalued.

    we can try exploit this if we knew when the market was going to move that day. of course this is a fool's game. i've long struggled and tried to predict using fundamental data.....u get burned.

    because we dont know when the large movement can occur, we can do only one thing.

    stay alive till the big move comes, we must design our position/lot sizes and our VaR risks accordingly to ensure that we can come up with an respectable drawdown.

    i would start this off by calculating my total risk using 10 straight(2 weeks) days of loss. if i lost 10 days straight in a row, what is the drawdown im willing to accept? i will be conservative and say 1%.

    if the market moves silently for the 10 straight days, i will lose 1%.

    now lets calculate the lot sizing and stop losses. take profits are executed at the end of the london session.

    Initial Capital: 100000
    total expected drawdown(10 days) : 1%

    1000 dollars in expected loss for 10 days. 100 dollars in expected daily loss.

    the stop loss can be configured to accept 300 dollars in loss in the single day.

    but how many points should the stop loss be?

    you have to figure this out by trial and error, i've come up with a number and still testing it.

    basically, because we are looking for relatively larger rewards than risk, it would make sense that we should protect against being whipsawed by using larger SL, while reducing the position size accordingly. we can be rest assured because we know our expected drawdown within reasonable confidence in the course of 2 weeks.


    the key attractiveness of this system is the the disproptionate reward:risk relationship. it is at least 2:1 or 3:1 for violent pairs like GBP.


    anyhow, i just want some feed back, and cotinue to test it.