The thesis
SilentDip rests on a tidy behavioural premise: a sharp price drop without a negative headline is usually noise, and noise reverts. A drop carrying genuine bad news, by contrast, tends to keep drifting lower as the market digests it. So the strategy insists on two conditions at once — oversold technicals (an RSI/price dislocation) and the absence of salient negative sentiment — then exits as the name mean-reverts toward its SMA20 or RSI normalizes.
The idea is grounded in recent literature: Clinch, He, Landsman & Li's November 2025 SSRN working paper "From Drift to Reversal", itself extending Chan's 2003 Journal of Financial Economics study of price reactions to news and no-news. That is a real, citable foundation rather than a curve-fit hunch, which raises the prior that the edge is structural.
Backtest and validation
Over 1,279 trading days and 294 trades, the backtest returned 52.62% (final equity $15,262 from $10,000), an 8.69% CAGR, Sharpe 1.03, and a shallow 10.38% max drawdown. The 70.55% win rate fits a mean-reversion profile — many small wins, occasional giveback.
The validation is the strongest part of the story. Across 4 walk-forward folds, all 4 were positive, with performance improving over time: fold Sharpes of 0.37 and 0.35 in the choppy 2021–24 windows, then 1.84 and 1.91 in 2024–26. The most recent out-of-sample fold returned 16.75% with a 2.39% drawdown. Probabilistic Sharpe (PSR) of 0.991 says the positive Sharpe is almost certainly real.
The risks
Two numbers deserve scrutiny. First, the deflated Sharpe (DSR) of 0.563 — adjusting for the 42 trials run during development — is far more sober than the headline 1.03, and sits only just above a coin-flip. The edge survives the multiple-testing haircut, but not comfortably. Second, the early folds were barely profitable (1.91% and 4.96%); the flattering aggregate leans heavily on the last two years.
Turnover of 4,192% signals a lot of churn, though flat fees kept costs at just $294 and FX at zero in this universe of 24 US large-caps.
Live so far: mostly waiting
The live paper run began 2026-10-13 at $10,000 and now sits near $9,981 — essentially flat. Of six scheduled runs, five executed nothing; the lone trade was a buy of 29 BAC at $51 on 10-14. The book holds ~$8,516 in cash, roughly 85% uninvested.
That is the strategy behaving as designed: selectivity means long idle stretches when no clean no-news dip appears. The flip side is real cash drag and a live sample far too small to confirm anything. SilentDip has earned a live slot on the strength of its validation — now it needs qualifying setups to actually show up.