the fade writeup: how do you filter out the days it kills you?

Been chewing on the strategy-library piece about fading extended moves since before the open. Notes: - The honest part is the asymmetry. 1.2% target vs 2% stop. That's stated up front, which I respect, but I still can't get my head around living with it. - Math I did on a napkin: with those numbers you need roughly 63% winners just to break even before fees. Fine. Win rate on this style usually looks better than that. Until one day it doesn't. - The conditions (below the 300 average, volume floor, falling streak) all filter for "extended". None of them filter for *why* it's extended. A gap-and-go liquidation and a slow grind down look identical to that ruleset. My actual question for people who run mean reversion intraday: - Do you add a regime filter on top? Something like "only take the fade if the daily trend isn't already broken" or a volatility ceiling? Or does that just shrink the sample to nothing? - The time exit (1200s) — is that doing more work than people think? Feels like the real risk control, more than the -2%. - I premarket everything. Tempted to just blacklist names that already gapped hard and let the rest run. Anyone tried that or am I inventing extra rules to feel safe? Not trying to talk anyone out of the category. Just want to know what people actually bolt on before they let it trade unattended.