The thesis
RSI Snap-Back is a mean-reversion play on the seven largest US tech names — AAPL, MSFT, NVDA, GOOGL, AMZN, META and TSLA. The logic is clean: large-caps that overshoot to the downside tend to snap back, so the strategy buys the most oversold names at RSI below 35 and exits into strength above RSI 70. A hard four-slot book caps concurrent exposure and forces the strategy to rotate rather than accumulate. It is a disciplined, rules-first design — exactly the kind of constrained system that is easy to reason about and hard to let run away from you.
Recent activity
The live tape has been quiet. Across the six scheduled runs from 2 to 9 November, only one produced a fill: a 10-share NVDA buy at $204.04 on 2 November, following a 7-share GOOGL entry at $280.28 on 30 October. Every other session logged 0 executed, 0 rejected — the market simply isn't offering RSI-below-35 entries right now, which is consistent with a reversion engine that waits rather than forces trades.
That patience shows in the book. Cash has been pinned at $5,956.92, meaning well over half the account is uninvested and only two of the four slots are filled. Total equity has drifted from $10,022.55 on 4 November down to $9,751.11 on 9 November — the account is currently below its $10,000 starting line.
Backtest and validation
The backtest is the strategy's stronger card. Over 451 days it returned 20.95% (final equity $12,095.14), an 11.21% CAGR, with a 66.67% win rate across 37 trades. Fees were a trivial $37 and FX cost zero. Those are respectable numbers for a long-only reversion system.
The risk profile is more sobering. The Sharpe ratio of 0.61 is modest — returns came with meaningful volatility — and the 23.73% max drawdown is steep for a $10k book, a reminder that "snap-back" entries sometimes catch falling knives before they bounce. Turnover of 773% is also high: this strategy churns its capital nearly eight times over, so execution quality and slippage matter more than the tiny backtested fee line suggests.
The most important gap is what's missing: validation is null. There is no out-of-sample or walk-forward result on record, so the 20.95% figure should be read as in-sample until proven otherwise. A high win rate on a single historical pass is precisely where overfitting hides.
Verdict
RSI Snap-Back is a well-governed idea with an attractive hit rate and genuine discipline in its four-slot book. But a lumpy drawdown, heavy turnover, no validation pass, and a live account quietly underwater all argue for watching before trusting. The structure is sound; the out-of-sample proof is the work that remains.