In 1994, a small team from Perth led by Rich Swannell began designing Elliott Wave computer programs for traders. Swannell was a programmer first and trader second. Very few in the world of trading are good at both. You can find this information useful. http://www.investopedia.com/university/advancedwave/elliottwave4.asp
Everything works until it doesn't-the way to test a system is to try and break it-you need to eyeball the bad trades and you will see that nothing works most of the time. Try a single Ma- buy when it moves above,and vice versa
From my point of view, it seems kind of self-evident that the market moves in waves, nonetheless, I have always viewed Elliott Wave Theory in a light similar to the sentiment expressed above, and the same is true to a lesser extent when it comes to harmonic patterns and Fibonacci ratios. What I’m hoping to do starting this week is to define price waves in accordance with my own answers to two key questions: (1) Where is priced likely to initiate a new trend; and once I know where to look, (2) how do I establish that the process of forming a new trend has actually begun to take place? The answer to the first question is a matter of determining something I call “statistical support/resistance” or how far price is willing to separate itself from what I view as its unadulterated trajectory (or its overall average direction as measured using the most representative intervals within a given time frame) before it can no longer resist the pull to return to more typical “degrees of separation.” When I notice price has reached one of these levels (I use three), I answer the second question (I determine whether the process of forming a new trend has actually begun to occur) by relying on a moving average that is not so short as to return frequent “false positives” or “head fakes,” yet not so long that it lags behind the trend to such a degree as to seriously compromise its validity. There is no wiggle room with this method. The deviation levels are precise and the moving averages are specific. There are no extensions or alterations. The system will either work or it won’t. Charts will either fit perfectly or not at all. Though not the greatest example, I executed the above AUDJPY trade simply for illustrative purposes. If it actually pans out I will find it very encouraging since I’m making the trade under less than optimum circumstances simply as an example. Price is falling below my designated MA (not pictured) after having breached the first of three levels of statistical resistance (represented by the three red diagonal lines). (P.S. One of the advantages of such a system is I know immediately if things go awry, so when price reversed direction back above my MA, I instantly exited the position to lock in the four pips of profit I still had at the time.)
Does Elliott Wave Theory actually work? There is no way to verify that it does or doesn't because the theory is not based on any kind of rigorous scientific or engineering principals. No mathematical rigor is ever presented. It's just predicated on an observation that market prices look wave-like, or cyclical in nature. If you must model financial data as a wave-like signal plus additive white noise, for god's sake use DSP (digital signal processing) methods. At least DSP has a solid scientific and engineering foundation, and mathematical rigor. Just be aware the DSP was developed for use with physical systems, and financial systems are sentient.
I had to hop off the wave on the way down, but I’m hoping to be successful in riding it back up! (Note how price bounced off my second level of statistical support right to the pip!) (P.S. A second benefit to my system is that I know when there is good reason to exit a position. Price began to stall around 84.64, so I checked my charts and saw that this was a level of minor statistical resistance, so I went ahead and took the six pips profit I had, even though it was five pips away from my target, because, mathematically speaking, it was the prudent thing to do.)
At the risk of offending someone I've never seen traders have much luck with EW or other variations and permutations of wave count or predictive market cycle software on real markets on a real portfolio. If someone knows otherwise by all means please let us know. YMMV