The timeframe was expanded to 72 days. That allows three 72 day periods that starts on 3/31/2020. It doesn’t make sense to look at the Feb-Mar panic numbers. There is some doubt about how much sense recent numbers make, but Feb/Mar 2020 numbers have to be approaching total worthlessness.
The daily Octal signal is x6. I decided to change the signal from Octal (Base 8, x0 -x7) to Quaternary (base 4, x0-x3). That implies two instead of three true false tests. That results in CO worth x2 and OC worth x1 if profitable. If both are profitable the signal is x3, and if neither is profitable x0. That makes the 8-Feb Quaternary signal = x2.
The top table of the graphic highlights how x1 and x2 returns for CO and OC are going over the previous 72 trade days. If the signal is x1, CO tends to be profitable and vice versa. The two lower tables show performance if CO is positive or negative the next day. For x2, CO is losing money overall. If CO loses, OC tends to be more positive. If CO wins, OC tends to be less positive.
I decided to call the new daily signal Kwat as an alternative to Quat. Since the logic diagrams are Kabbalistic (based on the sefirotic tree structure discussed on SeekingAlpha), there was the idea of calling the concept Quantitative Qabbalah, and then this might be Quant Quat Qabbalah. Rather than getting into Q so much, I thought staying with K might be better. But Kwant kwat kabbalah (KKK) has its own issues.
In any case, the difference of opinion between xFF and kwat x2 was hardly enough to convince anyone to sell on Friday’s OC mild weakness.