The Thesis
RSI Snap-Back is a mean-reversion play on the Magnificent Seven — AAPL, MSFT, NVDA, GOOGL, AMZN, META and TSLA. The idea is straightforward and well-worn in the literature: large-cap tech names that get stretched to short-term momentum extremes tend to revert. The strategy buys the most oversold names when RSI falls below 35, exits when RSI pushes above 70, and caps itself at a tight four-slot book. That hard limit is the discipline mechanism — it forces rotation into the cheapest-by-RSI names and bounds how much concurrent drawdown the book can carry.
It's a clean, legible strategy. You can explain it in a sentence, and the four-name ceiling means it never quietly drifts into an over-concentrated or over-levered posture.
Recent Activity
The notable thing about the last two weeks is how little happened. Across six scheduled runs between 23 and 30 September, every single one reported 0 executed, 0 rejected — cash $10,000, total $10,000. The book is empty and fully in cash.
This isn't a malfunction; it's the strategy doing exactly what it says. If none of the seven names is sitting below RSI 35, there is nothing to buy, so it waits. September's tech tape evidently never handed it an oversold entry. The flip side is that a strategy earning nothing while parked in cash is also earning no return, and a long stretch of inactivity is a reminder that entry signals this specific can go quiet for weeks at a time.
Backtest & Validation
Over 451 days the backtest returned 20.95% (final equity $12,095), an 11.21% CAGR, with a 66.67% win rate across 37 trades. Costs were negligible — $37 in fees, zero FX.
The quality numbers are more sober. The Sharpe of 0.61 is modest, and the 23.73% max drawdown actually exceeds the total return, meaning an investor would have had to stomach a drop larger than their eventual gain to collect it. Turnover of 773% is high for a four-name book — this strategy trades a lot relative to its size, which in a higher-fee or higher-spread environment would erode the edge faster than the $37 here suggests.
The most important caveat is what's missing: validation is null. There is no out-of-sample or holdout record attached. A 66.67% win rate on 37 trades is an encouragingly small sample to hang confidence on, and without a validation pass we can't distinguish a genuine reversion edge from a backtest fit to one favourable 15-month window of Mag-7 price action.
Verdict
RSI Snap-Back is honest about what it is — a disciplined, low-cost mean-reversion bet that trades rarely and sits in cash when signals are absent. The return is real but the risk-adjusted picture (sub-1 Sharpe, drawdown above return, no validation) argues for patience over conviction. The current all-cash stretch is a feature, not a worry — but it's also a reminder to judge this one over a full cycle, not a quiet fortnight.