The thesis
PanicFade v4 makes a specific, falsifiable bet: negative headline shocks tend to be overreactions, and prices revert toward fundamentals once the panic clears. It buys names where sentiment drops below -0.4 and the tape already confirms distress — RSI14 under 35 and price below its SMA20 — then sizes into the weakness. Exits are equally mechanical: it takes profit when RSI14 reclaims 50 or price recovers the SMA20, and hard-stops on fresh lows under continued negativity, treating that combination as a genuine fundamental break rather than noise.
The design is deliberately conservative. It fades panic only in a curated set of 24 large caps (AAPL, JPM, XOM, WMT and peers), uses live-quote-only cash accounting, and applies sanity-banded stops. This is a strategy built to avoid catching falling knives in illiquid or structurally impaired names.
Recent activity
The live account has been quiet — which is the thesis working as intended. Of the last six scheduled runs, five executed zero trades; the strategy simply had no qualifying panic to fade. The one exception was 15 September, when it bought 21 shares of BAC at $59.86, following a 28 August entry of 10 shares of WMT at $103.115. As of the latest run, cash sits at $7,696 against a total of roughly $9,987 — meaning only about a quarter of capital is deployed, and the account is hovering marginally below its $10,000 starting line. Live performance is, so far, essentially flat.
Backtest and validation
This is where PanicFade v4 earns attention. Over 1,233 days it returned 37.93% (final equity $13,793), a 6.79% CAGR, with a Sharpe of 0.96, a 62.88% win rate across 530 trades, and a contained 9.3% max drawdown.
More persuasive is the validation. All four walk-forward folds were positive, and — notably — performance improved over time: 1.6%, 3.88%, 10.14%, then 21.7% in the most recent fold, with that fold posting a 2.44 Sharpe. The probabilistic Sharpe ratio is 0.985, and the deflated Sharpe ratio, which penalises for the 27 trials run, still lands at 0.56. Clean out-of-sample behaviour and honest deflation accounting are exactly what you want to see before trusting a backtest.
Strengths and risks
The strengths are genuine: a coherent, literature-grounded thesis; robust, monotonically improving validation; and disciplined risk control. The risks are just as real. The 6.79% CAGR is modest, and turnover of 5,807% is punishing — 530 trades means fees and slippage matter more than the frictionless backtest implies. A DSR of 0.56 is respectable but not commanding. And the live account, flat and lightly deployed, has not yet demonstrated the edge out of sample in real time. PanicFade v4 looks statistically sound; the open question is whether the market will hand it enough panic to fade.