# Direction Pt. 1

It’s taken me a while to work on my next “panel” for Direction. I mean at the end of it all, if you can’t get direction correct, you’ll never win. Thus I don’t expect to get anything near as precise as I was with the projections. At least for now. Further, it can be a bit of a challenge to identify “direction”. What counts as an “up” day? C-O>0? H-O>O-L? Position of the end of the daily “wave”?

Per my short conversation with Bryan I decided I’d try to break down the fill %s a little more. I broke price into 4 components: H-O, O-L, H-C, C-O. Two measurements to add to Omega, one anchored on Open, the other to Close. Is there some pattern that exists should you examine the Fill%s of each of the components, rather than just relying on one (like I did in previous statistics)? Makes more sense to do so. Keep in mine that with my discoveries of Fill % previously, Fill %’s above 100 are just as likely (if not slightly favored) to Fill%’s under 100. Thus even assuming 50-50, you’d have overfills and underfills all over the place, which makes it difficult to examine. I put on a filter to see if there was something more noticeable if the Fill %s I was looking at were more extreme.

(green over fills, red under fills)

What do I want? Where is the trade opportunity? I believe it makes most sense to try to deduce where either the next H/O or O/L Overfill will occur to long or short on the open. I then reasoned that what I care about is first starting with locating under fills, if it has any bearing on future overfilling to “make up” lost fill.

I created my first “trajectory” which is either long or short bias, and which I think is not very good at all. I’m struggling to find a good clean simple way to define the short term “trajectory”/bias. Then, combining the two, I looked at what happens when the H/O overfills/underfills when the trajectory is long, and likewise what happens when the O/L overfill/underfills when the trajectory is short.

If logic follows, HO overfills on the long trajectory and OL Overfills on the short trajectory will show no edge. Looks good. The HO underfill on long trajectory actually looks decent- 60% (is that 140% considered the bump? perhaps skew is not great enough). However, the OL underfill is not showing enough of an edge.