I Coded Powell Trades' Strategy | Pt. 1
About this lesson
https://quantpad.ai/ Powell Trades has a video called "The setup that made me 60k in February" where he walks through one of his strategies. In this video, I do a good-faith programmatic implementation of it in Pine Script and put it through a proper validation pipeline for prop firm deployment. The build: daily zones he calls rejection blocks set the directional bias, a five-minute fibonacci retracement extends off the reactionary leg, and a five-minute entry trigger inside the discount half of that retracement fires the trade. Stop-loss goes past the far edge of the entry trigger, take-profit at the nearest viable five-minute pivot. I walk through every assumption I had to make where the rules in his video were ambiguous or contradicted themselves on camera (anchoring the fib differently between two examples in the same video, for one), build the strategy module by module so you can see exactly where the discretionary fog gets quantified, then run an out-of-sample optimized regime filter on it before pushing it through a reshuffling Monte Carlo on a TopStep $150K account. The verdict: positive net expected value on prop firm account challenges — ~22.4% pass probability, ~$206 net EV per account after fees, ~10 days mean time to first payout — driven by a low win rate with asymmetric winners. Technically profitable on funded account challenges. Not something I would deploy with live capital. The full Pine Script, the Python regime filter optimizer, and the backtests are all uploaded to the QuantPad social library. Edit the rules, retest it, and tell me where the implementation differs from how you read his video. 🔬 Topics covered: top-down bias state management, fibonacci retracement entry logic, range-based intraday zones, regime filter optimization, reshuffling Monte Carlo, prop firm pass probability, funded account challenge expected value, TopStep $150K account modeling. 📐 Tools used: Pine Script, TradingView, Python (regime filter optimization), QuantP
DeepCamp AI