the time exit in that sleep-safe crypto setup bugs me a bit

read the strategy library post about running crypto unattended with three separate ways out. the part that stuck with me is the reasoning for writing the exits as separate branches instead of one big condition — "clarity is a safety feature". i've definitely written combined conditions at 1am that i could not decode a week later, so fair. but i keep chewing on the HoldTime exit. 2400 seconds, and it fires regardless of where the position sits. so if i'm at +1.8% and the clock hits, out i go, one notch under the take profit. that's the whole point of a time stop i guess, it doesn't negotiate. still feels like it throws away the good tail along with the dead ones. two things i want to ask people who actually run stuff overnight: 1) does anyone make the time exit conditional, like only cut on time if ProfitPct is between say -0.5 and +0.5, and otherwise let the other two branches handle it? or does that just quietly reinvent the combined condition the author was warning about 2) the angle scaling note. the same degrees over a longer window demands a much bigger move, so widening the window means lowering the angle. i normally run 1m-5m stuff and i was going to stretch the window out to calm down the entries. so i'd be dropping the threshold at the same time. Which means i'm changing two knobs at once and i'll have no idea which one did what. anybody got a sane way to test that one variable at a time without spending a month on it not planning to leave anything running until i can explain both branches out loud sober.