Optimisation gets an improvement but not quite up to the original. This is odd  because the ATR trail works better almost everywhere else I’ve tried it, and the reenter should help. I tried a compromise of conventional break even with ATR trail but was worse. Attached image is with these values/changes:

//if dayofweek<5 then
ctime= hour>1
//elsif dayofweek=5 then
//ctime= hour>1 and hour<t2
//endif





// breakeven stop atr
once breakevenstoptype     = 1    // breakeven stop - 0 off, 1 on

once beatrperiod    = 10    // atr parameter value
once beminstop      = 12    // minimum breakeven stop distance

once bepointsorperct= 1     //[0]percentage [1]points

if tradetype=1 then
once bestoplong     = 5   // ts atr distance
once bestopshort    = 5   // ts atr distance
bepointstokeep      = 3  //-(sl/2) // positive or negative //not use once







once trailingstoptype     = 1    // trailing stop - 0 off, 1 on

once tsatrperiod    = 10         // ts atr parameter
once tsminstop      = 12         // ts minimum stop distance

if tradetype=1 then
once tsincrements = .25         // set to 0 to ignore tsincrements
once tsminatrdist = 1
once tssensitivity        = 0    // [0]close;[1]high/low
elsif (tradetype=2 or tradetype=3) then
once tsincrements = .25         // set to 0 to ignore tsincrements
once tsminatrdist = 1
once tssensitivity        = 1    // [0]close;[1]high/low
endif

if trailingstoptype then
if barindex=tradeindex then
if tradetype=1 then
trailingstoplong     = 8   // ts atr distance
trailingstopshort    = 8   // ts atr distance
elsif (tradetype=2 or tradetype=3) then
trailingstoplong     = 6   // ts atr distance
trailingstopshort    = 6   // ts atr distance
endif

Any other changes worth trying? Fifi’s pivots are inactive i think but I’m not sure how to bring them into play

DOW-5m-Mother.of_.Dragons-v1p.jpg