RSI (Relative Strength Index): How It Works & What Indian Traders Watch For

What is the Relative Strength Index?

The Relative Strength Index (RSI) is a momentum oscillator developed by J. Welles Wilder and introduced in his 1978 book New Concepts in Technical Trading Systems. It measures the speed and magnitude of recent price changes to evaluate whether a security has been gaining or losing value at a rate that is historically unusual.

Unlike moving averages, which are trend-following (lagging) indicators, RSI is an oscillator — it fluctuates between 0 and 100 and is plotted in a separate panel below the price chart. It does not overlay the price itself; it provides a normalised reading of momentum that can be compared across different securities and time periods.

RSI is one of the most widely used technical indicators globally and in India. It appears as a default indicator on most charting platforms including those used for NSE and BSE analysis. For a brief definition, see our RSI glossary entry.

The RSI formula: step by step

The RSI calculation involves three steps. The standard lookback period is 14 (14 days on a daily chart, 14 weeks on a weekly chart, etc.):

Step 1: Calculate price changes.For each of the last 14 periods, compute the change from the previous period's close:

Change = Close today − Close yesterday

Separate the changes into gains (positive changes) and losses (absolute value of negative changes). If a period had no change, both gain and loss are zero for that period.

Step 2: Calculate average gain and average loss. For the initial 14-period calculation:

Average Gain = Sum of gains over 14 periods / 14 Average Loss = Sum of losses over 14 periods / 14

For subsequent periods, Wilder used a smoothing method (similar to an exponential moving average):

Average Gain = (Previous Avg Gain × 13 + Current Gain) / 14 Average Loss = (Previous Avg Loss × 13 + Current Loss) / 14

Step 3: Calculate RSI. First, compute the Relative Strength (RS):

RS = Average Gain / Average Loss

Then normalise it to a 0-100 scale:

RSI = 100 − (100 / (1 + RS))

If Average Loss is zero (the security gained every period), RS is undefined and RSI equals 100. If Average Gain is zero, RS is zero and RSI equals 0. In practice, RSI rarely reaches these extremes.

Reading the 0-100 scale

The RSI scale has conventionally been divided into zones:

• Above 70: Overbought zone. The security has experienced sustained upward momentum over the lookback period. The average gain has been significantly larger than the average loss. This does not mean the price is about to decline — it means the pace of gains has been historically elevated.
• 30-70: Neutral zone. RSI is in a range where neither gains nor losses have dominated to an extreme degree. Most of the time, RSI fluctuates within this band.
• Below 30: Oversold zone. The security has experienced sustained downward momentum. Average loss has dominated average gain. Again, this does not automatically mean a reversal is imminent.

Some analysts adjust these thresholds depending on the market regime. In strong bull markets, RSI may spend most of its time between 40 and 90, rarely touching 30. In bear markets, it may oscillate between 10 and 60, rarely reaching 70. Andrew Cardwell, who extended Wilder's work, proposed using 80/40 in uptrends and 60/20 in downtrends as more appropriate thresholds.

RSI divergence: bullish and bearish

Divergence occurs when the RSI line moves in the opposite direction to the price chart. There are two types:

Bullish divergence. Price makes a lower low (a new trough below the previous trough), but RSI makes a higher low (its trough is above the previous trough). This has historically been interpreted as an indication that downward momentum is weakening even though price continues to decline. The logic is that sellers are losing force — each new low is produced with less selling intensity than the previous one.

Bearish divergence. Price makes a higher high, but RSI makes a lower high. This has historically indicated that upward momentum is fading even as price continues to rise — each new high is produced with diminishing upward force.

Historical example.During the Nifty 50's rally from the March 2020 COVID lows through October 2021, several instances of bearish divergence were observed on the weekly RSI chart as the index made successively higher highs while RSI formed lower highs. Eventually, the index corrected from its all-time high in October 2021. However, earlier bearish divergence signals during the same rally had not led to significant reversals — the trend had continued upward for months despite the divergence. This illustrates that divergence signals can persist for extended periods and do not have a reliable timing component.

Important caveat. Divergence is one of the most misused concepts in technical analysis. The presence of divergence does not guarantee a reversal. It is a condition that has preceded some reversals, but it has also appeared without any meaningful reversal following. It is best viewed as a secondary confirmation tool rather than a primary signal.

RSI as a trend strength indicator

While RSI is commonly taught as an overbought/oversold indicator, its more reliable use — as suggested by Wilder himself and later by Constance Brown and Andrew Cardwell — is as a trend strength gauge.

In a strong uptrend, RSI tends to oscillate between 40 and 90. Dips toward 40-50 represent pullbacks within the trend, not oversold conditions. In a strong downtrend, RSI tends to oscillate between 10 and 60. Rallies toward 50-60 represent relief bounces within the downtrend, not overbought conditions.

This framework changes how RSI is interpreted: instead of looking for reversals at 70 and 30, the analyst watches for the range within which RSI is oscillating to determine the prevailing trend regime. If RSI consistently finds support near 40-50 and peaks near 70-80, the trend is upward. If RSI finds resistance near 50-60 and troughs near 20-30, the trend is downward. A shift in this oscillation range can signal a trend change before a crossover is visible on the price chart.

Failure swings

Wilder described the "failure swing" as one of the strongest RSI-based signals. There are two types:

Bullish failure swing. RSI drops below 30 (into oversold territory), bounces above 30, pulls back but stays above 30, then breaks above the level of its previous bounce. This pattern occurs entirely on the RSI chart — price is not directly part of the signal. It has historically been observed before some significant reversals from downtrends.

Bearish failure swing. RSI rises above 70 (into overbought territory), drops below 70, rallies but fails to exceed 70 again, then breaks below the level of its previous dip. This has historically preceded some downward moves.

The failure swing is distinct from divergence in that it is a self-contained RSI pattern that does not require comparison with the price chart. Wilder considered failure swings more reliable than divergence, though both are subject to false signals.

RSI on different timeframes

RSI can be applied to any timeframe — from 5-minute intraday charts to monthly charts:

• Intraday (5-min, 15-min). RSI on intraday charts generates frequent signals. Indian intraday participants have used RSI on 5-minute and 15-minute charts with shorter lookback periods (7-9) for faster reads. However, noise levels are high, and false signals are more common.
• Daily. The standard timeframe for RSI analysis. A 14-day RSI is the default on most platforms and the most commonly referenced in Indian financial media.
• Weekly. Provides a smoother, higher-conviction reading. Weekly RSI overbought/oversold signals have historically been associated with more significant turning points, though they are also less frequent.
• Monthly. Used for long-term structural analysis. Monthly RSI reaching extreme levels on a broad index like the Nifty 50 has historically been rare and significant.

Limitations of RSI

RSI is a powerful tool, but it has well-documented limitations:

• Prolonged overbought/oversold readings.In strong trending moves, RSI can remain in overbought or oversold territory for weeks or months. Historically, Reliance Industries' RSI stayed above 70 for extended periods during its major rally from 2017 to early 2018 as the stock more than doubled. Participants who treated the overbought reading as a reversal signal would have exited prematurely.
• False divergence signals. Divergence can persist for extended periods without a reversal. Bearish divergence during a strong uptrend may simply reflect a deceleration of momentum rather than an impending reversal. Acting on divergence alone without confirmation has historically led to premature entries.
• Sensitivity to the lookback period. Changing the RSI period from 14 to 9 or 21 can produce significantly different readings and signals on the same chart. There is no universally optimal period — the choice involves a trade-off between sensitivity and reliability.
• Not useful in isolation. RSI provides one dimension of information — momentum. It does not account for volume, trend structure, support/resistance levels, or fundamental factors. Using RSI as the sole basis for analysis has been consistently identified as a limitation in the technical analysis literature.

Combining RSI with other indicators

RSI is most commonly used in combination with other tools:

• RSI + moving averages. Moving averages identify the trend direction; RSI identifies momentum extremes within that trend. For example, in an uptrend confirmed by price above the 200-day SMA, an RSI dip toward 40-50 has historically been watched as a potential continuation setup.
• RSI + support and resistance. An oversold RSI reading near a known support level provides two independent data points suggesting the same conclusion — that the area may attract participation. Neither is reliable alone, but confluence has historically increased the quality of signals.
• RSI + volume. Rising volume during an RSI divergence signal has historically been viewed as strengthening the case for a potential reversal. Low volume during divergence has weakened it.
• RSI + candlestick patterns. A bullish reversal candlestick pattern (such as a hammer or morning star) forming when RSI is oversold has historically been considered a higher-quality signal than either component alone.

Practical considerations for Indian markets

Indian equity markets have characteristics that affect RSI interpretation:

• Gap openings. Indian stocks frequently open with significant gaps due to overnight global cues (US market close, SGX Nifty, Asian markets). A large gap can cause RSI to spike or drop sharply within a single period, creating extreme readings that may not persist. This is particularly relevant on stocks with lower domestic liquidity.
• Event-driven volatility. Quarterly earnings announcements, RBI policy decisions, and Union Budget sessions create sharp, event-driven moves that can push RSI to extremes. These event-driven readings reflect a specific catalyst rather than a sustained momentum shift, and historical behaviour after such events has varied widely.
• Large-cap versus mid/small-cap. RSI signals on Nifty 50 constituents have historically been smoother and more interpretable than on mid-cap or small-cap stocks, where lower liquidity and higher volatility create noisier RSI readings.

Related tools and further reading

To understand how price interacts with key levels where RSI signals gain context, see our Support and Resistance guide. For visual pattern recognition on price charts, see our Candlestick Patterns guide. For reading price charts from first principles, see How to Read Stock Charts.

This article is educational only and does not constitute investment advice, a trading signal, or a solicitation to transact in any security. RSI readings are derived from historical price data and do not predict future price movement. Overbought and oversold labels are descriptive, not prescriptive. Past market behaviour is not indicative of future results. Please consult a SEBI-registered investment adviser before making any trading or investment decision.