Filters & KISS methods

Discussion in 'Strategy Building' started by grainmerchant, Oct 11, 2003.

  1. I think simple is better. Here is a simple entry filter for daytraders.

    If price is higher than yesterday's high, take only long entries that the system signals.

    If price is lower than yesterday's low, take only short entries that the system signals.

    If price is neither above or below the previous range a) take no entries or b) take only oscilliation type entries.


    Are there any major problems with that type of thinking?
    What are some simple exit filters/techniques that you adhere to?
     
  2. Quah

    Quah

    I have only one rule:

    Go long - sell when it goes up - if it doesn't go up, don't go long.
     
  3. balda

    balda

    can you define "goes up" please.
     
  4. Everybody says keep it simple. It cracks me up. I have found nothing about trading simple.
     
  5. balda

    balda

    in my understanding "simple" is when you do not use 10 indicators for entries or exits.

    something like 15 min time frame is in up trend and on 5 min time frame two indicators give you buy signal.
     
  6. I'd say my strategy is pretty simple, but executing it perfectly without hesitation or emotions takes a lot of discipline. You can look at my journal for info on that (link in sig).

    So, now, I'm not looking around for a "strategy" or some magic indicator. I'm just working on myself; on the psychological part of trading which is actually tougher than finding a strategy.

    -Fast
     
  7. I realize that the usual meaning is : dont overcomplicate things, but I have found that trading 1 minute flags with no indicators is not a simple matter. There are a many, many things to be considered. If it was simple we'd all be rich.
     
  8. Quah

    Quah

    Goes higher than your entry point. If you follow my rule, you'll never lose.
     
  9. prophet

    prophet

    There is no easy answer.

    There are plenty of common sense, simple strategies, enhancements, and publicly available systems that are net profitable but perhaps have undesirable risk, deep drawdowns, or only work in certain markets.
    -> Easy to find, high risk

    There are simple strategies that can be successfully improved through simple methods such as performance monitoring, stops, filters, hedging, inter-market diversification or combined with other systems or fundamental analysis.
    -> Harder to find, less risk

    There are sophisticated strategies that fail because they are too complex, “curve fit” past market action and can’t generalize future action, or lead to excessive volatility in returns.
    -> Easy to find, high risk.

    There are sophisticated strategies that succeed because they are internally diversified, “lateral” in design, either profiting from multiple market inefficiencies/patterns or a few patterns well conserved over time, but hidden nonetheless. Adaptive systems can work too as long as the adaptive elements are lateral, robust and well monitored and controlled.
    -> Very hard to find, lower risk.

    There is everything in between these extremes. Millions of traders, investors and intelligent agents (adaptive systems) are searching for low risk systems and strategies (mechanical or discretionary). The more a strategy becomes profitable and well known, the more it is traded, the more it’s execution prices worsen. Alternatively, more sophisticated participants may catch on to a popular system and recognize that the ones trading it are risk adverse, then step into the markets to fade it, purposely causing others to hit their stops or get squeezed, causing predictable breakouts or mean reversions. Again, a popular strategy becomes less profitable. Sometimes there are cycles where the markets move from one phase to another. Periods of uncertainty, tight ranges and stagnation alternate with periods of trending, greed and fear. Often there is a continual evolution of what works and doesn’t. In my experience a low risk system requires extensive study of markets, either in real-time or through back testing and forward testing systems. One needs to find a novel strategy not already being traded and thus not subject to being overloaded or faded. Real money trading is important too. However it’s also possible to learn about markets and develop a system through papertrading alone, as long as one properly accounts for slippage, or have traded enough to understand size, liquidity and slippage factors.
     
  10. T-REX

    T-REX

    I say we demonstrate "LIVE" here on ET the rationale behind these various market timing systems.

    I have a live simulation on Tuesday regarding the "2 Bar Charlie Method"!

    Why doesnt somebody simulate their methodology as well. That way we could either prove or disprove of the many myths regarding trading systems in realtime live!

    :D :D :D

    Then we will see just how complex or simple some of these BS methods "REALY ARE" and we will see if they "REALLY WORK".
     
    #10     Oct 12, 2003