The Thesis
RSI Snap-Back is a mean-reversion play on the seven names that dominate the market's attention: AAPL, MSFT, NVDA, GOOGL, AMZN, META, and TSLA. The premise is straightforward and, frankly, well-worn in academic literature: large-cap tech tends to overshoot on short-term momentum extremes and then revert. The strategy buys the most oversold names on RSI below 35, exits into overbought territory above 70, and enforces a hard four-slot book so that no more than four positions can accumulate drawdown at once.
The design choice worth highlighting is that discipline is structural, not discretionary. A capped book limits concurrent exposure and forces rotation rather than conviction-stacking. That is a sensible guardrail for a reversion system, where the temptation is always to average down into a name that keeps falling.
Recent Activity
Here the picture is quiet — arguably too quiet. Across six scheduled runs from August 11 through August 18, the strategy executed zero trades and rejected zero candidates. It has been sitting in a full $10,000 cash position the entire window, with total equity unchanged.
This is not a malfunction; it is the thesis working as specified. If none of the seven names printed RSI below 35, there was simply nothing oversold enough to buy. A mean-reversion system is supposed to wait for dislocation, and a market grinding sideways or higher gives it nothing to do. Still, an idle live book means the on-record performance is entirely backtest-driven for now — worth keeping in mind before reading too much into the headline numbers.
Backtest and Validation
Over 451 days the backtest returned 20.95% total (11.21% CAGR), ending at $12,095 on a $10,000 start. The win rate is a respectable 66.67% across 37 trades, and fees were negligible at $37 with no FX drag.
The caveats deserve equal billing. The Sharpe ratio is a modest 0.61 — this is not a smooth ride, and the risk-adjusted return is unremarkable. Max drawdown reached 23.73%, meaning nearly a quarter of the book was underwater at the worst point; the four-slot cap limits breadth but clearly not depth. Turnover of 773% signals a high-churn approach that leans heavily on execution assumptions holding up in live conditions.
Most important: the validation field is null. There is no walk-forward or out-of-sample confirmation on record. A 66% win rate and a 20% return on a single historical path are encouraging, but without out-of-sample testing we cannot rule out that the RSI thresholds are tuned to this particular window. For a strategy already trading live, that gap is the first thing I would want closed.
Verdict
RSI Snap-Back is a clean, disciplined expression of a credible idea, and its patience — sitting in cash rather than forcing trades — is a feature, not a bug. But the case is not yet proven: a middling Sharpe, a real drawdown, high turnover, and no validation run mean this belongs in the promising-but-unconfirmed bucket. The next milestone is a walk-forward pass and enough live trades to see whether the backtest survives contact with the market.