What does PF=1.075 tell us?

Discussion in 'Strategy Building' started by trd, Aug 4, 2009.

  1. trd

    trd

    It's IMO a wrong theory :)
    Can you calculate the annual lower and upper bound expectancy of a stock
    with a current price of $100 and with an annual historical volatility 0.35 ?
    Ie. statistically worst and best expectancy after a year.
    My calculations give Low=$70.47, High=$141.91
    So, if you buy and hold for a year then be prepared to lose 29.53% !
    Ie. you are risking 29.53% of your money.
    Now do the same calc for smaller time frames...
    Conclusion: the shorter the timeframe the less is the risk percent. Cf. the result I had posted previously (it has only -1.5% MDD risk for the whole year of giving back to the market from current account value...)
    Compare this with the 29.53% above... :)
    High Freq Trading is the way to go as it limits the risk substantially.
    Of course one must do it right... :)
     
    #11     Aug 5, 2009
  2. trd

    trd

    Ok, agreed :)
     
    #12     Aug 5, 2009
  3. Retief

    Retief

    Before you quit your day job and/or commit real money to a system with a profit factor of less than 1.1, you might want to test your theory for a month or two with a simulation account. Keep a journal here and show everyone that profit factor is almost worthless as a parameter for evaluating the profitability of a trading system. It should be interesting.
     
    #13     Aug 5, 2009
  4. Some posts are painful to read...

    slippage? commissions? trader error? risk? backtesting often means almost nothing? etc...

    Profit Factor is one of the most important stats there is. Unlike some others, it is hard to manipulate when you have a large enough sample size.

    1.075 says "sorry, try again..."
     
    #14     Aug 5, 2009

  5. Why you think that PF is worthless ?




    P.S. I have to agree with TZ that PF is very valuable
     
    #15     Aug 5, 2009
  6. It means that you have to do a butt load of transactions per month if you want a steady monthly check.
     
    #16     Aug 5, 2009
  7. jim c

    jim c

    What if you have a PF of 1.1 but you are hitting 85 % of your trades? Is PF used alone worthless....but means a little more when used in conjunction with % profitability?
    Where do you draw the line with PF vs profitability? Jim
     
    #17     Aug 5, 2009
  8. I think the weakness of PF is that it does not include transaction costs.

    You can have PF of 2 but due to A LOT of transactions > marginal profitable
     
    #18     Aug 5, 2009
  9. winning% does not mean anything. It is a stat that newbies cling to.

    PF means how much $$$ did I win over how much $$$ did I lose? Most of what WP tells you, is contained in PF.

    A knowledgeable trader will subtract ALL commissions and slippage from "wins" if doing any kind of backtest.

    Further, they will also subtract something for lapses - occasional errors (long instead of short), forgetting to place a stop, overleveraged, IP connection went down, broker error, etc.

    And further, since trading is abut Risk-Adjusted Reward, you should really deduct for Risk also. You can always blow out under certain situations.

    You also need to adjust for the fact that few backtests carry forward in the same way. Few.

    So 1.075 is basically breakeven to losing.

    I would consider anything 1.4 and under as not profitable. And preference for at LEAST 1.7 or higher.
     
    #19     Aug 5, 2009
  10. jim c

    jim c

    So a system that wins 85% of the time with a 1.1 PF is the same as a system with 1.1 pf and a 50% win %? I do not agree. Win % is NOT conatined in PF. It just isnt. If you win 1.00 for every 1.00 lose...and you win 85 % of the time....seems to me you would make money?
    the question Im trying tp understand is...where do draw the line. Is it 65 % wins with a PF of 1.7 just as an example? Jim
     
    #20     Aug 5, 2009