The thesis
PanicFade v4 bets on a well-documented asymmetry: negative headline shocks tend to be overreacted to, then revert toward fundamentals. The rules are disciplined rather than heroic. It buys large caps only when three conditions coincide — sentiment below −0.4, RSI14 under 35, and price beneath its 20-day SMA — sizing into the panic rather than waiting for confirmation. Exits are equally mechanical: it takes profit on reversion (RSI14 back above 50 or a reclaim of the SMA20) and hard-stops fresh lows under continued negativity, treating that as a genuine fundamental break rather than noise.
The universe is 24 blue-chip names across tech, financials, healthcare, staples, and energy. That matters: fading panic works far better in liquid, well-covered large caps than in speculative small caps, where a sentiment shock is more likely to be correct.
Validation
This is where PanicFade earns attention. The strategy passed a 4-fold walk-forward test with all four folds positive, and — notably — performance improved out of sample rather than decaying: fold returns ran 1.6%, 3.9%, 10.1%, and 21.7%, with the final fold posting a standout 2.44 Sharpe. The probabilistic Sharpe ratio (PSR) of 0.985 is strong, indicating the full-sample Sharpe is very likely above zero.
The caveat is the deflated Sharpe ratio (DSR) of 0.56 across 27 trials. Once you penalise for the number of variants tested, confidence drops to merely respectable — this is a strategy that survived selection, not one that dominated it. The improving fold pattern is encouraging but could also reflect a favourable regime in 2025–26 rather than durable edge.
The numbers
Over 1,233 backtest days the strategy returned 37.93% (final equity $13,793 from $10k), a 6.79% CAGR, with a 62.9% win rate across 530 trades and a contained 9.3% max drawdown. The headline Sharpe is 0.96 — decent, not spectacular. Turnover is heavy at 5,807%, though fees stayed modest ($530 total, no FX cost), so the churn isn't eroding returns the way it might elsewhere.
Recent activity
Here the picture cools. Since the lone WMT buy on 28 August (10 shares at $103.12), the strategy has sat on its hands: six consecutive scheduled runs from 7–14 September each executed zero trades and rejected zero, cash parked at $8,958.54 against a total account value drifting narrowly between roughly $10,016 and $10,055. That is the thesis working as designed — with markets calm and no name hitting the panic thresholds, PanicFade correctly does nothing. But a strategy that only trades in dislocations will show long dead patches, and investors should expect flat, uneventful stretches punctuated by bursts of activity.
Verdict
PanicFade v4 is a credible, well-specified contrarian engine with a clean validation record and sensible risk controls. The risks are honest ones: a modest deflated Sharpe after many trials, a single-digit CAGR that rewards patience over excitement, and a signal that stays dormant for weeks at a time. It is a strategy to judge over dislocations, not calendar quarters.