Discussion in 'Strategy Development' started by mightymight, Jul 20, 2006.

  1. anybody get really frustrated because their automated strategies dont seem to hold up over long periods of time, anyone have insight into developing working strategies and the mind frame needed . . .
  2. Why not just use the ones that have been working for 100 years. Use pivot and daily high and low. 10,20, 50 moving averages, macd, double tops and bottoms, head and shoulders. Why complicate things?
  3. Mighty - the reason this happens is simple - markets change and a computer 'set it and forget it' system cannot respond to market conditions and changes. This is why you do not see blackbox systems last for extended periods of time. In the end, the markets are influenced by humans (which are not predictable) and computers (which thrive on predictability) cannot respond.

    Just my 2 cents. Abandon the automated system idea and learn to trade live, by yourself.
  4. Qwerty


    Quote from mightymight:

    anybody get really frustrated because their automated strategies dont seem to hold up over long periods of time, anyone have insight into developing working strategies and the mind frame needed . . .


    The markets are dynamic, the underlying cause is the dynamic struggle that is occurring between bulls & bears. The reason that the markets move up & down. An automated system cannot update the condition of the bulls & bears up to the second, nor can markets be bound by a variable.

    As long as a trader fails to understand(interpret) this battlefield
    between bulls & bears, true understanding of why markets move will always remain elusive.

    An automated strategy is admirable, however, expect guaranteed frustration in the near future. Try to focus your attention on price, nothing is above price. Make copies of charts consisting of different time frames & undertake a comprehensive
    study of them.
  5. This is something I have been pondering. I want to backtest to develop probabilites, but what sample is too small or large?

    I believe some systems may have to large of a sample and over optimized and experience serious drawdown at times. You may start trading the auto system at the time it experiences it's worst drawdown ever. Will you have the intestinal fortitude to keep going? Most don't. I didn't. Black boxes can't handles breaking news.....terrorism, Bernake making dumb comments, and the like IMO. That is their major flaw.

    The more I learn about this venture, the more I believe that you have to have a soundly backtested and forward testing strategy and manually trade it while making sure you have your mind straight. These three things will beat any black box IMO.
  6. Perhaps there would be no frustration if we don't expect any automated system will last forever. :confused:
  7. what is forever in the black box world? a week, month, year, etc I suppose its all relative and if we knew the answer we wouldnt have to ask and we would all be very wealthy.
    In response to many of your comments, why then do many people trade the same strategies, technicals, charts etc for many years and find them to be profitable systems. They find their indicators and they stick to them for good or bad and it seems they do this rather than deciding what to use moment to moment, and constantly changing their mind which leads to them missing all the signals. Whats the difference between a disciplined trader and a black box?
    Thanks for the response all.
  8. The successful disciplined trader can use his discretionary skills to know when to step up his size, reduce it, cut out early, hold on longer etc depending upon the current factors. I suppose the blackbox has limited criteria that it can use to find entries and therefore isnt as flexible?

    Every trade setup, regardless of how similar it is to past trade setups, will be slightly different in some ways. And these differences are what the disciplined trader sees and reacts and adjusts to.

  9. Just 2 cents:

    Probably no robust system, even an adaptive one which depends on the number of look-back period, is without some limitations or restrictions in relation to even-changing market characters/ events/ shocks.

  10. Tums


    you can re-invent the wheel, or you can start with Van Tharp.
    both route takes investment in time, one costs less in money and effort.
    #10     Jul 25, 2006