I was thinking that the separation between #1/#2 and between #7/#8 would also be important, especially when detecting a “change” of the best position.
For example, the difference for 5-Apr of #1 (AUD) to #2 (USD) is 0.08% using Dennis’ numbers.
Looking at this graphically, I see that AUD, USD, and CHF are so close together that having one switch the high position wouldn’t be all that relevant. In fact, using my broker, CHF does take the #1 slot, which it keeps for the first H4 this morning, but AUD takes it back the next H4.
- my values differ from Dennis’ because I use a different broker.
- my dates are one ahead of Dennis’ because I use server time.
- I’ve un-normalized JPY so I can see what it’s doing relative to other currencies.
So, what separation would be meaningful? Maybe 0.15 to 0.20% minimum. I’m open to other interpretations.
Here is some data if anyone wants to try it out.
SW Data with Separations.csv (99.6 KB)