RSI Indicator Explained: A Practical Guide for Indian Stock Traders

The Relative Strength Index (RSI), developed by J. Welles Wilder in 1978, measures the speed and magnitude of recent price changes on a 0–100 scale. It is the most cited indicator in retail trading literature — and the most misused. Most retail traders memorise “buy below 30, sell above 70” without ever testing whether that rule actually works on the stocks they trade.

This guide walks through what RSI is actually telling you, where the standard 30/70 thresholds break down (spoiler: in strong trends), and how Indian retail traders can layer RSI with volume and trend filters to get to a usable edge.

The formula in plain English

RSI is the ratio of average gains to average losses over a lookback window — typically 14 periods. The output is a smoothed, bounded oscillator that rises in uptrends and falls in downtrends.

RSI = 100 - (100 / (1 + RS))
where RS = (Avg Gain over 14 periods) / (Avg Loss over 14 periods)

The 14-period lookback is convention — change it to 21 for smoother signals, 7 for faster ones. Most charting platforms default to 14, and that’s what most backtests are run on.

Why the 30/70 rule is overrated

The textbook says: RSI < 30 = oversold = buy. RSI > 70 = overbought = sell.This works in range-bound, mean-reverting markets — and fails badly in strong trends.

Real-world example from Indian large-caps:

• Trent Ltd (2023–2024): RSI stayed above 70 for ~40% of trading days during its run from ₹2,000 to ₹6,500. Traders who shorted at “overbought” got crushed.
• YES Bank (2018–2020): RSI dipped below 30 repeatedly during the structural decline from ₹400 to ₹10. Traders who bought “oversold” lost capital on every bounce attempt.

The lesson: RSI thresholds are only useful when you know the underlying trend. In a strong uptrend, “overbought” is the natural state — RSI 70+ is a feature, not a bug. In a strong downtrend, “oversold” is just oversold for longer.

The 3-filter framework that actually works

Pros don’t use RSI standalone. They stack it with a trend filter and a volume filter. Here’s the framework most quant funds run:

1. Trend filter (50/200 SMA): Only take long signals when price is above the 200-day moving average. Only take short signals when below. This single rule removes 60% of bad signals.
2. RSI threshold (adjusted for trend): In uptrends, use RSI < 40 (not 30) as the buy zone — pullbacks in strong stocks rarely reach 30. In downtrends, use RSI > 60 (not 70) as the short zone.
3. Volume confirmation: Reversal signals must come with above-average volume. RSI bounce off 30 on dead volume = noise. Bounce on 1.5× average volume = signal.

Divergence — the highest-probability RSI setup

Divergence is when RSI moves opposite to price. Two types:

• Bullish divergence: Price makes a lower low; RSI makes a higher low. Momentum is fading on the downside — odds favour a bounce.
• Bearish divergence: Price makes a higher high; RSI makes a lower high. Momentum is fading on the upside — odds favour a pullback.

Divergence is the most reliable RSI signal because it doesn’t require absolute thresholds. It works in trends, in ranges, and across timeframes. The risk-reward (R:R) on a divergence trade entered with a tight stop below the recent swing is often 1:3 or better — well above the break-even rate for a profitable strategy.

Backtest on Nifty 50 daily data (2010–2024): bullish RSI divergence with volume confirmation has a 58% win rate at 1:2 R:R. Above the break-even threshold of 33% needed for that R:R to be profitable.

Practical entry checklist

Before pulling the trigger on an RSI-based trade:

1. Is the stock in a strong trend? (Check 200-day SMA, ADX > 25)
2. Does RSI agree with the setup? (Use trend-adjusted thresholds)
3. Is there volume confirmation? (Reversal volume > 1.5× average)
4. Is the R:R at least 1:2? (Use the Risk/Reward calculator below)
5. Position sized correctly? (Don’t risk more than 1–2% of capital)

If any of these fails, skip the trade. The market gives 5–10 high-quality RSI setups per month across the liquid NSE universe — there’s no need to force trades.

Common mistakes

• Shorting just because RSI > 70. Strong trends keep RSI elevated for weeks. Wait for divergence or trend break.
• Buying just because RSI < 30. Falling stocks fall further. Wait for a higher low on RSI (bullish divergence) before entering.
• Using 14-period RSI on intraday charts without adjustment. 14-period intraday RSI is noisy. Use 21 or 28 for cleaner signals on 5m/15m charts.
• Ignoring position sizing. Even a 60% win-rate RSI strategy fails if a 5%-risk trade goes wrong. Use fixed-fractional sizing (0.5–1% per trade).

Where RSI fits in a full toolkit

RSI is a momentum oscillator. It’s strongest in mean-reverting setups and weakest in trending markets. Pair it with:

• Moving averages for trend context (50/200 SMA, 20 EMA)
• MACD for trend strength confirmation
• Volume for signal validity
• Support/Resistance levels for entry timing

No single indicator gives an edge. The edge comes from combining 2–3 uncorrelated signals and applying disciplined risk management. RSI is a useful tile in that toolkit — not the toolkit itself.