The RSI Mean Reversion Strategy

A Number That Measures Exhaustion, Not Direction

The RSI Mean Reversion Strategy uses the Relative Strength Index, an oscillator that compares the size of recent gains to recent losses and expresses the result as a single reading between zero and one hundred. (The Technical Analysis → Indicators section explains exactly how RSI is calculated; this page assumes you already understand the basic mechanics and focuses purely on trading it.) A reading above roughly seventy is conventionally called overbought, and a reading below thirty is called oversold.

The strategy's core idea is that momentum, like a stretched rubber band, tends to snap back toward a more moderate level after moving too far too fast in one direction. When a genuinely good stock, already in an established uptrend, suffers a sharp short-term pullback that pushes RSI into oversold territory, that pullback often represents a temporary overreaction rather than a change in the underlying trend, and frequently offers a lower-risk entry point for a trader wanting to join that trend.

The critical, non-negotiable qualifier in that description is the established uptrend. RSI oversold readings occur constantly in falling stocks too, and buying those readings without a trend filter is not mean reversion trading; it is catching a falling knife wearing a mean-reversion costume. This page builds the version that avoids that trap.

Why the Trend Filter Is Not Optional

The single most important design decision in this strategy is the mandatory trend filter, and it deserves its own explanation before the rules themselves. RSI measures how fast and how far price has moved recently, and nothing about that calculation distinguishes a healthy pullback within a genuine uptrend from the early stages of a serious, structural decline. Both can push RSI equally deep into oversold territory.

The difference between the two is entirely a matter of the broader trend, which RSI itself does not measure. A stock trading well above its 50-day or 200-day moving average that suffers a sharp pullback is behaving consistently with ordinary trend behaviour, a pause within an intact uptrend. A stock already trading below those same averages that keeps making new oversold readings is, more often than not, simply continuing to fall, and each new oversold reading in that decline has historically preceded further declines nearly as often as a genuine bounce.

This is precisely why the strategy is taught long-only, on the long side, applied strictly to stocks already trading above a rising longer-term moving average. RSI oversold readings in a confirmed downtrend are excluded from this strategy's rules entirely, not because they never bounce, but because the odds of a genuine, sustained bounce are meaningfully worse there than inside an uptrend.

Setting the Right RSI Thresholds for Indian Large Caps

The conventional thirty and seventy thresholds work reasonably well as a starting point for large, liquid Nifty 50 constituents, which tend to have more moderate, less extreme momentum swings than smaller, more volatile names. Some traders tighten the oversold threshold to twenty-five for particularly stable, low-volatility large caps, since these names rarely reach even thirty without a genuinely sharp pullback.

More volatile mid-cap and small-cap stocks, in contrast, can spend extended periods below thirty during a healthy, ongoing uptrend, simply because their day-to-day price swings are naturally larger. Applying the standard thirty threshold unmodified to these names can generate premature entries well before the actual short-term low. Loosening the threshold toward thirty-five for genuinely volatile names, or requiring RSI to turn back upward from below thirty rather than entering the moment it crosses the threshold, both help adapt this strategy to a wider range of Indian equities.

The Playbook, Rule by Rule

Every rule below exists to keep this strategy inside genuine uptrends and out of genuine declines.

• Setup condition: Price trades above a rising 50-day moving average on the daily chart, confirming an established uptrend, exactly as required by the Moving Average Pullback Strategy on this site.
• Trigger: The 14-period RSI drops below 30, or below 35 for a more volatile mid-cap or small-cap name, signalling a short-term oversold condition within the confirmed uptrend.
• Entry: Enter once RSI begins turning back upward from below the oversold threshold, rather than the moment it first crosses below it, to avoid entering while the short-term decline is still accelerating.
• Stop-loss: Place the stop below the most recent swing low, or one Average True Range below entry (see the Indicators section for ATR), whichever gives a tighter, better-justified risk relative to the pullback's own structure.
• Exit: Close the position, or take substantial partial profit, once RSI recovers back above 50 to 55, signalling the short-term oversold condition has resolved; trail the remainder using the 21-day moving average if the broader uptrend remains intact.
• Position size: Risk one to one and a half percent of trading capital on the distance between entry and the stop-loss placed at the pullback's own low.

Note that this strategy's exit target is deliberately modest: a recovery back to a neutral RSI reading, not a fresh overbought extreme. This strategy is built to capture the snap-back from an oversold extreme toward the middle of the range, not to ride the entire subsequent trend leg, which is a different job better handled by the trend-following strategies on this site.

A Bounce on the Hindustan Unilever Chart

Illustrative example built to show the mechanics, not a specific historical trade record.

Consider an illustrative example on Hindustan Unilever, a stable, low-volatility FMCG large cap that has historically shown fairly clean RSI mean-reversion behaviour within its longer uptrends.

The stock has traded above a clearly rising 50-day average for three months, advancing steadily from around 2,380 to 2,540. A sharp two-session pullback, triggered by broader market weakness unrelated to the company itself, pushes the stock down to 2,470, and its 14-period RSI drops to 27, a genuinely oversold reading for this typically stable stock.

Two sessions later, RSI turns upward, reading 33, while price holds near 2,478, confirming the pullback has not broken the broader uptrend. The trader enters near 2,485. The stop is placed at 2,445, below the pullback's own low, a risk of roughly 40 points. Over the following eight sessions, the stock recovers to 2,555, and RSI climbs back to 56, triggering the exit near 2,550.

The trade captured roughly 65 points on a 40-point risk, a reward-to-risk near 1.6 to 1, over a relatively short holding period of just over a week, consistent with this strategy's modest, snap-back-focused objective rather than an extended trend-following target.

Divergence: A Sharper Version of the Same Signal

An experienced version of this strategy adds one further refinement worth understanding even though it is not part of the core mechanical rules above: RSI divergence, where price makes a new short-term low but RSI itself fails to make a correspondingly lower reading. This pattern suggests the selling pressure driving the pullback is genuinely weakening beneath the surface, even as price itself continues briefly lower.

A bullish divergence occurring inside an already-confirmed uptrend, alongside an oversold RSI reading, has historically offered a somewhat higher-conviction version of this strategy's core signal than an oversold reading without divergence. Spotting divergence reliably takes practice and is genuinely subjective at the margins, which is why the core mechanical rules above do not require it as a mandatory condition, but traders comfortable reading it can use it as an additional confluence factor rather than a replacement for the trend filter.

Costs, Frequency and the India VIX Context

Because this strategy trades relatively short-lived oversold dips rather than sustained trends, it tends to generate more frequent signals across an active large-cap watchlist than the pure trend-following strategies on this site, which brings a correspondingly higher cost drag from brokerage, exchange charges, GST, stamp duty, and slippage. Position sizing and realistic reward-to-risk expectations, both smaller and shorter in duration than a trend-following trade, need to reflect that higher trading frequency honestly.

India VIX, the market's own volatility gauge, provides useful context for this strategy specifically. During periods of elevated India VIX, oversold RSI readings occur more frequently and can be somewhat less reliable, since heightened overall market volatility can push RSI into oversold territory on broader market fear rather than any genuine, stock-specific mean-reversion opportunity. Being more selective, or reducing position size, during periods of unusually high India VIX is a sound adaptation for this strategy.

What Gets Skipped in Most RSI Content Online

RSI is one of the most talked-about indicators in retail trading content, precisely because the number itself is simple to read. Here is what that simplicity usually leaves out.

• An RSI reading below 30 is not a fixed law; it is a convention. Some genuinely healthy uptrending large caps rarely dip below 35 even during real pullbacks, while volatile mid-caps can spend days below 30 without any real change to their broader trend, making a single universal threshold a rough guide rather than a precise rule.
• The trend filter matters more than the RSI reading itself. A trader forced to choose between confirming the trend properly and using the exact textbook RSI threshold should always prioritise the trend filter; it is the difference between this strategy and an entirely different, much riskier one.
• RSI divergence, when it is genuinely present, has historically offered a meaningfully higher-conviction version of this signal, yet it barely features in most beginner-level RSI content, which tends to focus purely on the overbought and oversold thresholds.
• This strategy's reward-to-risk ratio is structurally smaller than a trend-following trade's, because it targets a return to a neutral reading rather than an extended trend leg. Traders expecting trend-following-sized gains from a mean-reversion signal are measuring the wrong strategy against the wrong yardstick.
• The traders who use RSI mean reversion profitably over years are rarely the ones with the most finely tuned threshold. They are the ones who never once bought an oversold reading in a stock trading below its own moving average.

Before risking real capital, paper trade this exact rule set on ten signals and log every one, win or lose, in your Trading Diary. Patterns you cannot see while trading, you can see clearly while reviewing.