The Thesis
PanicFade v4 bets that negative headline shocks are overreacted to and tend to revert toward fundamentals. The entry rule is specific: buy names hit by sentiment below −0.4 that are already technically oversold (RSI14 under 35 and trading below their 20-day moving average), then size into the panic. Exits are symmetric and disciplined — take reversion when RSI14 climbs back above 50 or price reclaims the SMA20, and hard-stop on fresh lows under continued negativity, treating that as a fundamental break rather than a dip. The approach is grounded in recent sentiment-asymmetry literature (arXiv:2509.11970, 2025) and runs on a 24-name large-cap universe spanning tech, financials, healthcare, and staples.
Recent Activity
Activity has been quiet and selective. Over the last two weeks most scheduled runs executed zero trades, with bursts on 7 and 13 October. The 13 October run sold NKE (29 shares at $35.53) and HD (4 at $293.95) — both acquired on 21 September at $36.10 and $299.12 respectively, so these look like stop-outs at small losses, exactly the fundamental-break exits the thesis describes. Earlier, a UNH position bought at $370.31 was closed at $377.96 for a gain, and BAC was rotated in and out. The live paper account currently sits around $9,795 total with $7,434 in cash — meaning the strategy is holding a large cash buffer and is modestly underwater against a presumed $10k start. In other words, the live book has not yet reproduced the backtest's shine.
Validation and Backtest
This is where PanicFade v4 earns its "live" status. The backtest returned 37.93% over 1,233 days (a more sober 6.79% CAGR), with a 0.96 Sharpe, 9.3% max drawdown, and a 62.88% win rate over 530 trades. Crucially, the walk-forward validation passed cleanly: all four folds were positive, improving monotonically from 1.6% (Sharpe 0.19) to 21.7% (Sharpe 2.44). Out-of-sample return was 21.7% at a 2.44 Sharpe, and the Probabilistic Sharpe Ratio of 0.985 is excellent.
Strengths and Risks
The strengths are real: four-for-four positive folds, a shallow drawdown, and a high PSR are hard to fake. But balance demands three caveats. First, the Deflated Sharpe Ratio is only 0.56 across 27 trials — once you penalize for multiple testing, confidence drops sharply from the headline PSR. Second, turnover is enormous at 5,807%, so the thin $530 fee line flatters a strategy that trades constantly; real-world slippage could bite. Third, the live account is currently flat-to-negative, a reminder that a backtest's best fold was its most recent, and recency can mislead.
Verdict: a well-constructed, literature-grounded contrarian strategy with honest validation discipline. The oversold-plus-negative-sentiment filter is sensible, and the stop logic is working as designed. Watch the DSR and the gap between backtest and live before reading too much into the 37.93%.