The thesis
RSI Snap-Back is a mean-reversion play on the Magnificent Seven — AAPL, MSFT, NVDA, GOOGL, AMZN, META, and TSLA. The premise is simple and well-trodden: large-cap tech names that get stretched to a short-term momentum extreme tend to snap back. The strategy buys the most oversold names (RSI below 35), exits into strength (RSI above 70), and enforces discipline with a hard four-slot book that caps concurrent exposure. It runs live.
Backtest performance
Over 451 trading days the strategy turned a starting book into $12,095.14 in final equity — a 20.95% total return, or 11.21% CAGR. The win rate is a healthy 66.67% across 37 trades, which fits the mean-reversion profile: frequent, modest wins.
The less flattering numbers sit right beside them. The Sharpe ratio is 0.61 — positive, but well short of what you'd want before sizing up. Maximum drawdown reached 23.73%, meaning the equity curve gave back nearly a quarter of its peak at some point. And turnover ran to 773% over the period, an aggressive churn that the four-name rotation demands. Fees came to $37 with no FX cost, so friction was contained here, but a 7.7x turnover rate leaves little margin if slippage or spreads widen.
The validation gap
The single most important field in the record is empty: validation is null. A 20.95% return with a two-thirds win rate reads well until you notice there is no out-of-sample or walk-forward check backing it. Mean-reversion on a seven-name universe is exactly the kind of setup where in-sample results flatter — a handful of well-timed reversals on NVDA or TSLA can carry an entire curve. Without validation, treat the backtest as a hypothesis, not a verdict.
A conspicuously quiet live book
The live record is where the story turns cautionary. Across six scheduled runs — August 7 through August 14 — every single one reports the same line: 0 executed, 0 rejected, cash $10,000, total $10,000. The book is fully in cash and has been idle for at least a week. No trades have printed.
That is not necessarily a fault. A disciplined RSI-below-35 entry should stay flat when nothing is oversold, and a market where the Mag-7 is grinding sideways or higher simply won't trigger it. But it does mean the live strategy has yet to demonstrate the reversion edge its backtest claims. The capital is parked, earning nothing, waiting for a signal.
The balance
RSI Snap-Back has a clean, defensible thesis and a backtest that clears the bar for a second look. The four-slot discipline is a genuine strength — it hard-caps how much the strategy can lose to any one name. But the risks are equally clear: a modest Sharpe, a 24% drawdown, heavy turnover, and — most of all — no validation to confirm the edge is real rather than fitted. Until the live book actually trades and a walk-forward test fills that null field, this one stays firmly in the prove-it column.