The Thesis
RSI Snap-Back is a systematic mean-reversion strategy targeting the Magnificent Seven — AAPL, MSFT, NVDA, GOOGL, AMZN, META, and TSLA. The logic is straightforward: large-cap tech names, heavily owned and continuously repriced by institutional flows, tend to overshoot on short-term momentum swings and snap back. The strategy enters when RSI drops below 35 (oversold) and exits when RSI climbs above 70 (overbought), maintaining a hard cap of four concurrent positions. That four-slot constraint is a deliberate risk-management choice — it prevents the book from becoming a diversified long-only tech fund masquerading as a tactical strategy.
Backtest Performance
Over 451 days of paper trading, RSI Snap-Back grew a $10,000 portfolio to $12,095, a 20.95% total return and an annualised 11.21% CAGR. The win rate of 66.67% across 37 trades is respectable for a mean-reversion system — two out of three entries resolved in the strategy's favour.
The Sharpe ratio of 0.61 is the most honest number here. It reflects that returns were real but not smooth: a 23.73% peak-to-trough drawdown at some point during the period would have tested conviction meaningfully. For context, a drawdown of that magnitude in a $100k real-money account is a $23,700 paper loss at the worst point — not comfortable for most retail investors, even if the strategy eventually recovered.
Turnover clocks in at 773% — the book rotated through its full notional value roughly 7.7 times over the period. In a paper-trading context fees are nominal ($37 total), but real-money implementation would need to account for spread, slippage, and borrow costs on any leveraged version.
Recent Activity: A Strategy on Pause
The most notable current signal is what the strategy is not doing. Every scheduled run from July 3 through July 10 logged zero executions — the full $10,000 has sat in cash for at least six consecutive trading sessions.
This is not a malfunction; it is the system working as designed. RSI Snap-Back only fires when a Mag-7 name reaches extreme oversold territory. A week of inactivity implies the universe has been trading in a range that never broke below the RSI 35 threshold. Whether that reflects genuine market strength, low volatility, or a bull-tape that rarely offers the dips the strategy needs is worth watching. If this idle stretch extends, it raises the question of opportunity cost — cash sitting uninvested is a drag against any benchmark.
Risks and Open Questions
No out-of-sample validation. The validation field is currently empty. All reported metrics are in-sample backtest figures, which means they were computed on the same data the strategy was built against. Without a held-out test period or walk-forward analysis, there is real risk that the RSI 35/70 thresholds were fit — even unconsciously — to a specific market regime. This is the single most important gap before treating these numbers as predictive.
Regime dependency. Mean-reversion strategies on tech thrive in choppy, range-bound markets and struggle in strong trending phases (up or down). A sustained directional move in the Mag-7 — either a prolonged rally that keeps RSI elevated, or a structural drawdown where oversold keeps getting more oversold — could impair both entry frequency and exit profitability.
Concentration risk. A four-stock maximum from a seven-name universe means the book can be highly correlated. In a broad tech selloff, all four slots could be in drawdown simultaneously.
Bottom Line
RSI Snap-Back is a well-structured, rules-based system with a coherent thesis and a solid paper-trading track record. The win rate and total return are encouraging. The near-24% drawdown, high turnover, and — most critically — the absence of out-of-sample validation mean real-money deployment should wait for a proper forward test window. The current idle stretch is worth monitoring: if the strategy goes another two to three weeks without a trigger, revisiting the RSI thresholds against current market volatility regimes would be prudent.