I'm curious about something that I've noticed for quite some time now but never addressed. I have always developed rules that apply to both longs and shorts. In other words, the rules for longs are exactly the same as for shorts. That is the only way I've seen them written by other "professionals" as well. However, as I do all the programming that I do, I've seen rules that do well for one direction or the other but not necessarily for both over time and I mean a year of data. Right now, I have a system that is unbalanced to favor the short side for the last 13 months. Same rules for both. The longs is doing ok but not nearly as well as the shorts. So, I started playing with the rules for the longs to see if I could do something to improve performance and I have. Bear in mind that I'm testing over 13 months of data which includes 294 long trades and 250 short trades. So, statistically speaking, I think that's a pretty decent sample. That begs the question, should rules for long and short entries be written independent of each other or this this an indication of something bad?