Like impossible to test "this feels weak", but I always test as how I understand, also how someone else tests maybe quite differently to how I test, plus what data they are using like IB data or CQG. So many laugh and say stats are too much work to generate, but just tells me their amount of dedication to becoming good at the craft of trading.It's usually true break even, occasionally a tick or two loss, occasionally a tick or two gain. I prefer to get out with a limit order at B/E when a trade thesis is invalidated. So for example I enter with-trend on a pullback and price breaks down prior to testing the previous high or low. That's a sign of either a deeper pullback, possible range/chop, or even a reversal. So the breakdown signal tells me to move my profit target limit order to B/E and most of time I can get out that way before my stop loss is hit. With CT trades, though, which tend to have smaller profit targets, I'll actually move my stop loss to B/E after a favorable excursion clears a certain level. I'll also do this if price breaks out because a failed breakout can deteriorate rapidly.
All these decisions are based on statistics. I found quite some time ago that I can't trust thoughts like "that always happens" or "this feels weak" because once I applied statistical analyses to every occurrence of a particular setup (pattern + context + signal/entry method), I found that sometimes things I thought "always" happened only happened, say, 45% of the time!
I use Time stops, finding three bars best for my methods of day trading, either market moves my anticipated direction or if at a loss after three bars, my target become plus one tick. I can always get in again after if to me a clean cut signal says to enter but it is not to re-enter but giving me a new signal to enter. I find markets so much more choppy, it is easier to get the one tick, but that only after much testing first.
)