The thesis
PanicFade v4 bets on a well-documented behavioural pattern: negative headline shocks tend to be overreacted to, then revert toward fundamentals. The strategy buys names hit by sentiment below −0.4 that are already oversold on price — RSI14 under 35 and trading below their 20-day moving average — and sizes into the fear. It exits on reversion (RSI14 back above 50, or price reclaiming the SMA20) and hard-stops fresh lows made under continued negativity, treating those as a genuine fundamental break rather than a panic.
The design is grounded in recent sentiment-feedback literature (arXiv:2509.11970, 2025) and applied to a disciplined universe of 24 mega-cap names — Apple, Microsoft, JPMorgan, Walmart and the like. Confining a mean-reversion strategy to liquid large caps is a sensible guardrail: it reduces the risk of catching a genuinely broken small-cap on the way to zero.
Recent activity
The live paper account sits at roughly $10,040 in total value, having been quiet through early September. Six scheduled runs from the 8th to the 14th executed nothing — the market simply didn't hand PanicFade a qualifying setup. That patience is a feature, not a bug: a fade strategy that trades every day isn't waiting for panic.
The 15 September run finally fired, buying 21 shares of Bank of America at $59.86 and drawing cash down to about $7,696. The prior execution was a Walmart purchase on 28 August (10 shares at $103.12). Two trades in three weeks is consistent with a strategy that acts only on real dislocations.
Validation and performance
The backtest returns 37.93% total over 1,233 days (a 6.79% CAGR), with a 62.88% win rate across 530 trades and a contained 9.3% max drawdown. More importantly, the walk-forward validation is clean: all four out-of-sample folds were positive, and performance improved across them — the most recent fold (May 2025–Aug 2026) returned 21.7% at a 2.44 Sharpe with just a 2.6% drawdown. A probabilistic Sharpe ratio of 0.985 says the positive Sharpe is very unlikely to be noise.
The risks
Balance demands scrutiny of the softer numbers. The headline Sharpe is 0.96 — respectable, not exceptional — and the deflated Sharpe, which penalises for the 27 trials run during development, drops to 0.56. That gap is the honest signal: after accounting for the search, the edge is real but modest, and the standout final fold may flatter the whole.
Turnover of 5,807% is enormous, and while fees totalled only $530, that intensity makes the strategy sensitive to slippage and execution assumptions that a paper account understates. The recent BAC entry also concentrates a meaningful slice of equity into a single financial name.
Verdict
PanicFade v4 is a coherent, literature-backed strategy with a genuinely clean validation record — four-for-four folds is hard to fake. Temper enthusiasm with the deflated-Sharpe haircut and the turnover bill, and judge it on whether live results extend the out-of-sample trend rather than merely echo the backtest.