The thesis
RSI Snap-Back is a mean-reversion play on the Mag-7 — AAPL, MSFT, NVDA, GOOGL, AMZN, META, and TSLA. The premise is familiar and defensible: large-cap tech names tend to overshoot on short-term momentum and then snap back. The bot buys the most oversold names when RSI falls below 35, exits when RSI climbs above 70, and enforces a hard four-slot book. That last detail is the interesting one — the position cap is as much a risk-management device as an alpha source, deliberately limiting concurrent drawdown exposure and forcing the strategy to rotate rather than sprawl.
Recent activity: all quiet
Here is the headline for the live book: nothing is happening. Across its six most recent scheduled runs — from 2026-07-13 through 2026-07-20 — every single one reports the same line: 0 executed, 0 rejected, cash $10,000, total $10,000. The strategy is sitting entirely in cash and has not opened a position in that window.
This is not necessarily a failure. A mean-reversion system that only fires on RSI < 35 is, by design, supposed to wait. If none of the Mag-7 names have printed a sufficiently oversold reading, the correct behavior is to do nothing. The discipline is arguably the point. But it does mean prospective followers should understand this is a patient strategy that can go long stretches contributing zero — and, right now, we have no live trades to evaluate it on.
Backtest and validation
The backtest is where the case is actually made. Over 451 days the strategy returned 20.95% (final equity $12,095 on a $10,000 start), a CAGR of 11.21%, with a 66.67% win rate across 37 trades. That is a solid hit rate, and fees were negligible at $37 total.
The risk numbers demand more caution. The Sharpe of 0.61 is modest — this is not a smooth ride — and the max drawdown of 23.73% is steep, nearly larger than the total return itself. Turnover ran to 773%, meaning the book churned through its capital roughly eight times; the four-slot rotation is active, and while fees stayed low here, that pace leaves less margin for slippage in the real world.
The most important caveat: the validation field is null. There is no out-of-sample or walk-forward test on record. Every number above comes from the same backtest the rules were presumably tuned on, and with only 37 trades the sample is thin. A 66.67% win rate on that few trades can flatter a strategy that hasn't been stress-tested on unseen data.
Verdict
RSI Snap-Back has a coherent thesis, sensible position discipline, and a backtest that clears a reasonable bar on return and win rate. But the drawdown is uncomfortable, the Sharpe is unremarkable, the validation is missing, and — for now — the live book is doing nothing at all. Worth watching; not yet worth conviction.