Inspired by Connors and Alvarez: RSI(2) Mean Reversion
NO EDGEIn uptrending markets, deep short-term oversold readings (2-period RSI) mark high-probability bounce entries.
How this run differs from the source: Exits at the next close unless still oversold, not on the first close above the 5-day average; long only, daily from 2016.
| THE STRATEGY | $SPXTR BUY & HOLD | |
|---|---|---|
| Growth a year (CAGR) | 3.7% | 15.1% |
| Return per unit of risk (Sharpe) | 0.28 | 0.88 |
| Worst fall (max drawdown) | -66.6% | -33.8% |
// GROWTH OF $1, LOG SCALE
Tested on years it never saw (from 2021): 12.6% a year against 15.0% for the benchmark; Sharpe -0.16 in the training years, 0.63 after.
After 25 bps of trading costs: -48.8% a year. EDGE OVER RANDOM
// DECADE BY DECADE
| ERA | STRATEGY | BENCHMARK |
|---|---|---|
| 2010s | -9.6% | 15.1% |
| 2020s | 12.5% | 15.2% |
// THE IDEA, AS PUBLISHED
Larry Connors & Cesar Alvarez, 'Short Term Trading Strategies That Work' (2008) · 2008
- Instrument above its 200-day MA
- Buy at the close when RSI(2) is below 5 (thresholds up to 10 also shown)
- Exit at the close when price closes above its 5-day MA; no stop
- Mirror short rule below the 200-day MA (RSI(2) above 95)
A real computation on the QuantGPT warehouse, walk-forward. A study across the whole market includes only part of the companies that later delisted from 1998 (survivors only before); a study on your own list of tickers is not affected. Gross returns, no costs except where stated. Past performance is not a promise. Findings, not advice. COMPUTED 2026-10-19 18:59 ON THE QUANTGPT WAREHOUSE. COMPANIES THAT LATER DELISTED ARE ONLY PARTLY IN THIS UNIVERSE (A FIX IS IN PROGRESS), SO THE RESULT LEANS TOWARD SURVIVORS. FUNDAMENTALS BY FILING DATE. NOT INVESTMENT ADVICE.