I agree with both of you
All I am thinking is that with delta used alongside diff as I am doing then new positions are only opening when the slope is the right direction AND accelerating. So in theory where there is a pullback against the position no more positions are entered until it is turning back accelerating in the right direction again. So perhaps running for the second half of each session (where price often retraces) may not be a big issue in this case and any pairs where price just keeps accelerating will still provide the benefit.
I am very interested to see whether just mtf CSS cross or mtf CSS diff with delta or HGI (or both even?) proves more consistent over these next few weeks of forward testing.
I must admit I was struggling a bit with the session times so have changed my VPS and set file to GMT. I have used EUR open and close for London rather than London and this gives each session 9 hours of trading for the respective currencies instead of restricting to just 3.
... Maybe this will be too much but it solves the problem of a basket going the wrong way without new trades opening to counterbalance in the dead time between shorter trading hours.
You do not have the required permissions to view the files attached to this post.