Strategy Library

Trend

RSI(2) pullback

A short-term rule from Larry Connors. In a market above its 200-day moving average, a short, sharp pullback tends to bounce, so you buy when the 2-period RSI is extremely low and exit on the first bounce above the 5-day average. It was built and tested on stocks and stock indices on daily bars; the short side mirrors it below the 200-day.

IntermediateStocks and stock indices in the source; the sources do not cover forex, gold, crypto or intraday chartsDaily barsD1
Use this strategyPractice it

Example 01

RSI(2) pullback on a chart

US100 · D1RSI(2) under 5uptrend, dip, bounce

RSI(2) at entry: 1.9200-day SMA: 17,591.13Entry, on the close: 18,160.00Exit, above 5-day: 18,260.00Sample candles, not market data

The rules

Setup
  • For longs, the close is above the 200-day simple moving average (below it for shorts)
Entry
  • Buy when the 2-period RSI is at or below 5 (or 10); sell short at or above 95 (or 90) below the 200-day
  • Enter just before the close, as Connors did, or at the next open
Stop
  • The original rule uses no stop. Plan your own before you trade it, such as a fixed stop or a time exit
Target
  • Exit on a close above the 5-day simple moving average (below it for a short)
Manage
  • Optional: wait for the 2-period RSI to climb back above 50 before entering

Common mistakes

  • Dropping the 200-day filter. Without it you are buying every dip in a downtrend.
  • Trading it with no risk plan. No stop in the original does not mean no plan in yours.
  • Moving it to other markets. It was tested on stocks and indices on daily bars. Test before you assume it carries over.
  • Tuning the thresholds. Adjusting 5, 10 or 95 until the past looks good is curve fitting.

When it's off

  • No trade when the close is on the wrong side of the 200-day average
  • Expect losses when price keeps falling after the signal; the first signal is often early

Words to know

RSI overbought and oversold
Wilder's RSI (14 periods by default) reads above 70 as overbought and below 30 as oversold. It measures the speed of recent gains against losses, and in a strong trend it can stay overbought for a long time.
Moving average
The average closing price over a set number of candles, drawn as a line; the exponential version (EMA) weights recent candles more than the simple one (SMA). Traders use it as a trend filter or as a moving level for pullbacks.
Mean reversion
The idea that price tends to come back toward an average after an extreme move.
Curve fitting
Tuning a rule to past data until it stops working on new data.

When you make it yours, I'll ask what the rules above leave open. Start with one of these:

  • Is your trend filter the 200-day simple average on daily bars, on the market you trade?
  • Which threshold do you use, 5 or 10, and did you fix it before looking at results?
  • Do you enter at the close or the next open, and does your journal record which?
Ask Edge in DT Terminal

From DT Academy

What testing exists

The source reports Connors' own tests favouring the 5 and 95 levels and no stops. Independent tests vary, and none was found on markets other than stocks and indices.

Before you trust a backtest

  • Many professional traders risk under 1% of their account on one trade idea, and the common guideline is 1 to 2%.
  • The more versions of a rule you try on the same data, the more likely a good backtest is luck. Decide your rules before you test.
  • Include spread and commission, and practice on dates you have not looked at, before you trust a result.

Sources