A little project I was working on to avoid studying for finals. Using LazyBear's RSX code for a smoother RSI, then taking the RSX of fib number lengths. Take the average of that, then the JMA of that from the same fib numbers. The average of that is then treated as the trend, take the average of the trend values from the main time frames, the script calls pretty far back so adding a W or M TF I think would throw the calculations off. Then I smoothed that value using the jma's to create the overall trend. I got the idea from Ehler's Empirical Mode Decomposition about identifying peaks and valleys and creating an average of that to create a range. The idea is that if the trend is above the Average Peak then it is a bull trend, less than the average valley it's a bear trend, in between it's ranging. It looks like it turned out alright, I'll be working on this idea of fractals a lot this summer to see if I can improve it or build something better off of the idea.