@Everyone subscribed to this thread: As i said in the title of this thread this method of trading is experimental, i am a bit of a perfectionist so even though I have been trading this system I have continued to look for ways to improve it (not over-optimize), and earlier today I came to a few realizations about the current system and found ways to improve it drastically, here is a list of all of the changes (from the top of post#1 to the bottom) and why I have decided to implement them:
Removed from system:
EMA (5), EMA (10), (zone jumping): The original idea for adding the EMA’s was to use a psychological support to measure the strength of potential reversals, for instance when i saw an indecision candle I would zoom in to the 1 hour chart and see where the strength was, to see if price would break through these imaginary supports and resistances, and that is how i came to the zone jumping idea, that by seeing price dance around these psychological levels we could better determine where price would go, and it was and still is a valid idea HOWEVER I realized earlier today that the reasons they work is not what i originally thought, Although they do work it over complicates the principal they are trying to exploit
I noticed that nearly every valid signal could also derive direction from a candle close to candle close basis, meaning the fourth 1 hour candle in every 4 hour signal had a lower close than the first candle (if selling) and a higher close from the first candle (if buying), this led me to wonder why even with the difference between pin bars and reverse pin bars this was a consistent signal
The fact is reverse pin bars and pin bars both signal indecision, the difference between them is the way they do it, a pin bar typically overextends in the direction of the current trend and retraces back near its open, a reverse pin bar is the opposite as it overextends in the opposite direction of the current trend and retraces back near the open of the candle, these two different ways lead to two complete different types of stories on the 1 hour chart, which made me think of elliot wave theory!
I probably wont be able to explain my reasoning behind this but i realized that the waves within each 4 hour pin bar could almost be compared to an inner trend within a signal, because each signal is an indecision candle we can assume that before the indecision signal there was a trend in place, and after the indecision candle there will be a trend in place (either the same trend or the opposite), this made me want to compare each individual one hour candle to a swing, and if the final swing in this period of indecision is higher than the first swing in the period of indecision that would mean a buy, if it was lower that would mean a sell, and after tedious back testing i found this to be slightly more accurate then what the system currently used and much simpler…
Notes: Originally the wick idea was a basic form of inner candle swinging so there is no purpose for it anymore as that is what the whole system is right now, also the need for candles to be not all touching the EMA’s is gone too as this is a more accurate filter anyway and will normally catch the breakout that would result from price riding the EMA’s
Added to system: Instead of using zone jumping to determine entries, all we do now is look for the indecision candles, once we have found one we zoom in to the 1 hour chart and look at the first and last one hour candle within the 4 hour signal and determine our entry by the direction the closes are travelling, for instance if the fourth candle has a higher close than the first than we buy, if it has a lower close than we sell
The system is much simpler now and much more effective, I’m glad i posted it though otherwise i wouldn’t have been able to refine it so well, i will edit the original post now and put up new examples