Slippage and Win/Loss Ratio

Discussion in 'Risk Management' started by gsmaster, Feb 12, 2008.

  1. gsmaster


    Hi Everyone,

    I am new here at elitetrader and I am very sorry if what I am about to ask has already been answered like 1000 times, this place is so big that maybe having colleagues answering my question or just pointing me out the threads would be easier.

    I have been working on my system and I have to say that statistically it looks great, it makes money... obviously the stats do not include slippage and comissions but since I am testing and I am using a simulator (which also simulates some slippage) I am noticing that slippage is a real pain in the butt.

    So I started assuming that once I have statistical results I would do some slippage deduction and this is what I am calculating:

    Assume I am trading the ER2 I count that for each trade I will have 2 ticks slippage and 0.5 in comissions , so around 2.5 ticks in transaction fees if I calculate that for a 100 trades that would make around 250ticks or 25points!!!! just in fees... and I am not being conservative because I am not even counting the mistakes one do make.

    Obviously once I subtract that to my Total Profit I am still positive but the Win/Loss ratio is badly beaten...

    If I do a MonteCarlo equity simulation I can see that the equity curves after comission and slippage are quite dissapointing...

    Since my experience is limited I am really looking for advice from experienced traders, I need a reality check, I need to know if I am expecting things right,,,

    My average win/loss ratio is around 2- 2.5 and I win around 40% of the time ..

    Another question that I have is the following, people talk always about the famous 2:1 ratio now I ask myself are they often referring to:

    (avgwin/avgloss) ratio?


    (avgdollarwin / avgdollarloss) ratio?

    both can be different yet positive...

    I really thank everyone who cares to answer my questions in advance for your great help and I am sorry if this topic have been discussed already..

    Thank you
  2. ronblack


    According to the formula derived by Michael Harris, best case scenario with an average win/average loss of 2.5 produces a profit factor equal to 1.67. I consider it a low profit factor.

    If the ratio drops to 2.0 then the profit factor drops to about 1.3 and you are not making any money after slippage and commissions.

    The two ratios you mentioned should produce the same number. A ratio is a dimensionless number so the value of the point multiplier does not matter because it multiplies both numerator and denominator.

    I think you should read the article by Michael Harris " Derivation of the Profitability Rule and its Application in
    the Discovery of Trading Systems Based on Price Patterns":

  3. gsmaster


    thanks for the heads up,,, i will check the article... other comments are more than welcome..

  4. Simulation results can not lie.
  5. ronblack


    Good point!

  6. gsmaster


    Thanks for the details....