7. Strategy 4 — RSI-2 Mean Reversion

Fading the short-term extreme

The first three strategies were continuation — they buy strength and follow the trend. This chapter switches to the second engine of swing trading: mean reversion. The RSI-2 strategy buys when a stock is short-term oversold and expects a snap-back. It is the most famous documented short-term mean-reversion system.

The idea

Markets overshoot. In the short run, a sharp sell-off often pushes a stock too far, too fast — and prices tend to snap back toward the average, even if the longer-term trend is down. The RSI-2 strategy captures these bounces.

The key instrument is the 2-period RSI (RSI-2) — an extremely short-term oscillator that flips to deeply oversold (near 0) very quickly on a sharp drop. The strategy, documented by Larry Connors and Cesar Alvarez in their book Short Term Trading Strategies That Work, buys those deeply oversold moments and exits a few days later on the bounce.

The rules

Universe and timeframe: liquid stocks (or liquid index instruments). Daily timeframe.

Step 1 — The long-term trend filter (critical). Only trade in the direction of the longer trend. The classic version: take longs only when the stock is above its 200-day MA (an uptrend). In a downtrend, the same setup is used for shorts (or the trade is simply skipped). This filter is why the strategy works — buying oversold bounces in an uptrend is high-probability; buying them in a downtrend is catching falling knives.

Step 2 — The trigger. Enter a long when RSI-2 closes below 5 (some use below 10), signalling a deeply oversold short-term extreme.

Step 3 — Exit. Exit when price closes above the 5-day MA (the classic rule), or after a fixed number of days (e.g., 2–5), or at a fixed small profit target. The trade is a short bounce, held only days.

Step 4 — Stop. A stop below the entry day's low (or a fixed small percentage), for the times the bounce does not come.

Position sizing: standard small risk per trade; the strategy trades frequently, so costs matter.

Historical context

Connors and Alvarez published extensive historical results for RSI-2 mean reversion across US equities, showing that a simple "buy RSI-2 < 5, exit above the 5-day MA" rule — crucially, filtered by the 200-day trend — produced a high win rate over many years. The mechanism is well understood: short-term liquidity-driven overshoots revert, and the trend filter keeps the trader from fighting a genuine downtrend. The pattern generalises: short-term extremes tend to revert, especially with the longer-term trend.

Framed honestly: historically, buying deeply short-term-oversold conditions in assets above their long-term trend has tended to produce frequent small wins — with occasional larger losses when the "bounce" turns out to be a breakdown, which the stop is there to cap.

A worked example (a liquid NSE stock)

  • A large-cap stock is in an uptrend, trading above its 200-day MA at ₹800.
  • Over three days it drops sharply to ₹740 on broad market weakness. Its RSI-2 closes at 3 — deeply oversold. Entry at ₹742.
  • The 200-day MA is still well below — the longer trend is intact; this is a dip, not a reversal.
  • Over the next three days the stock bounces and closes above its 5-day MA at ₹775. Exit at ₹775.

Result: entry ₹742 → exit ₹775 = +4.4% in a few days, a short, high-probability bounce captured with a defined stop and a mechanical exit.

Risks and limitations

  • Catching falling knives. Without the 200-day filter, this strategy buys every crash and loses badly. The trend filter is not optional — it is the strategy.
  • Frequent small losses. Bounces sometimes fail; the stop caps each loss, but the strategy can string together several small losers. It relies on a high win rate with the filter to stay net positive.
  • Transaction costs. RSI-2 trades frequently (often holding only days), so brokerage and STT on each round trip matter more than for slower strategies. Run it on liquid names.
  • False confidence in high win rates. A high win rate can hide a few larger losses; risk per trade must remain small and disciplined.

Summary

  • Buy RSI-2 closes below 5 (or 10), exit above the 5-day MA — held only days.
  • The 200-day trend filter is what makes it work; without it, you are catching knives.
  • Documented by Connors & Alvarez; short-term extremes revert.
  • High win rate, frequent small trades — costs and stop discipline matter.

Next: Strategy 5 — the range fade.