RSI Explained: How to Use the Relative Strength Index to Find Overbought and Oversold Stocks

RSI shows up on nearly every stock chart, in nearly every charting platform, and in nearly every "how to read a chart" tutorial - which is exactly why it gets misused so often. Traders learn "70 is overbought, 30 is oversold" as a rule, and then watch a stock they shorted at RSI 75 keep climbing for another three weeks. The indicator isn't broken. The rule, taken alone, was always incomplete.

The Relative Strength Index, developed by J. Welles Wilder Jr. in his 1978 book New Concepts in Technical Trading Systems, is a momentum oscillator - it measures the speed and size of recent price moves, not the price itself. Used correctly, it tells you something price alone can't: whether the move that's already happened was driven by unusually strong or unusually weak momentum relative to that stock's own recent history. Used as a mechanical overbought/oversold trigger, it produces some of the most common losing trades in technical analysis.

This guide covers how RSI is actually calculated, what overbought and oversold levels really mean (and don't mean), RSI divergence and the failure swing pattern, how to pick a setting for your timeframe, and the specific mistakes that turn a genuinely useful indicator into a trap.

What RSI Actually Measures

RSI is bounded between 0 and 100, and it answers one specific question: over the lookback period, have gains been larger than losses, or have losses been larger than gains, and by how much?

That's a narrower question than "is this stock going up or down." A stock can be in a clear downtrend and still show a high RSI reading during a sharp relief bounce, because RSI only looks at the relative size of recent up-moves versus down-moves within the window, not the overall direction of the longer trend. This is the single most misunderstood thing about RSI: it is a momentum gauge, not a trend indicator, and not a price target.

How RSI Is Calculated

The math behind RSI is straightforward once you break it into steps. Using the standard 14-period setting on a daily chart:

1. Separate each day's price change into a gain or a loss. An up day contributes to the gain column; a down day contributes to the loss column. Flat days contribute zero to both.
2. Average the gains and average the losses separately over the 14 days, producing an Average Gain and an Average Loss.
3. Divide Average Gain by Average Loss to get the Relative Strength (RS) ratio.
4. Scale RS into the 0-100 range: RSI = 100 minus (100 divided by (1 plus RS)).

If Average Gain is much larger than Average Loss, RS is a large number, and the formula pushes RSI toward 100. If Average Loss dominates, RSI is pushed toward 0. If gains and losses have been roughly equal in size, RSI sits near 50 - which is why a reading near 50 is generally read as "no clear momentum bias" rather than as a bullish or bearish signal by itself.

After the initial 14-day calculation, virtually every charting platform switches to Wilder's smoothing method: each new day's gain or loss is blended into the running average rather than fully recalculated from scratch, using a formula equivalent to an exponential moving average. This is why RSI moves smoothly from day to day instead of jumping erratically, and why RSI computed on the same stock across different platforms occasionally shows very slightly different values if one uses a simple average and another uses Wilder's smoothing.

Overbought and Oversold: Starting Points, Not Signals

The textbook levels are RSI above 70 for overbought and RSI below 30 for oversold. These thresholds are useful, widely watched, and worth knowing - but they describe how much momentum has already happened, not what happens next.

Why this matters in practice: in a strong, sustained uptrend, it is completely normal for RSI to push above 70 and stay elevated for weeks while the stock continues making new highs. A trader who shorts the first RSI reading above 70 in a genuine bull trend is fighting the trend, not trading with an edge - and this is one of the single most common ways new technical traders lose money. The same applies in reverse during strong downtrends: RSI can sit below 30 for extended stretches while the stock keeps making new lows.

A more useful framework, sometimes attributed to technical analyst Constance Brown, treats the overbought/oversold range as shifting with the trend rather than fixed:

The classic "sell at 70, buy at 30" rule works best in genuinely range-bound markets. In trending markets, watching where RSI fails to reach on pullbacks (does it hold above 40? does it fail to reclaim 60?) is often more informative than the absolute level itself.

RSI as Trend Confirmation

Beyond overbought/oversold, RSI's behavior around the midpoint - 50 - is a simple, often-overlooked trend filter. In a healthy uptrend, RSI tends to stay above 50 on almost every pullback, only dipping below it briefly if at all. If RSI starts closing below 50 on pullbacks that previously held well above it, that's a sign the trend's underlying momentum is weakening, even if the price chart still looks intact. The reverse pattern - RSI failing to hold below 50 during downtrend bounces - can flag early trend exhaustion on the downside.

RSI Divergence

Divergence happens when price and RSI disagree about direction, and it's one of RSI's most-watched signals:

• Bearish divergence: price makes a new high, but RSI makes a lower high than its previous peak. The rally is still happening on the chart, but the momentum behind it is weaker than it was at the prior high - a warning that the move may be running out of steam.
• Bullish divergence: price makes a new low, but RSI makes a higher low than its previous trough. Selling pressure is technically still pushing price down, but with less underlying force than before.

Divergence is a heads-up, not a trigger. It can persist for weeks before price actually turns, and plenty of divergences resolve by price simply pausing rather than reversing. It works best combined with a second confirmation - a break of trendline support, a volume shift, or a moving average cross - rather than traded on its own. This topic has enough depth (regular vs. hidden divergence, how to spot it consistently, how to build an alert around it) to warrant its own full breakdown - see RSI Divergence Explained for the complete strategy.

The Failure Swing

The failure swing is a lesser-known RSI pattern that doesn't require comparing RSI to price at all - it's entirely visible within RSI itself:

Bearish failure swing: RSI rises above 70, pulls back, rallies again but fails to exceed its prior high (staying below the previous peak), then breaks below the low point of the pullback. That break is the signal - RSI failed to make a new high on the second attempt and then broke down.

Bullish failure swing: the mirror image below 30. RSI falls below 30, bounces, falls again but fails to make a new low (staying above the previous trough), then breaks above the high point of the bounce.

Wilder considered the failure swing one of the more reliable RSI patterns specifically because it doesn't depend on comparing RSI to price action at all - it's a standalone pattern within the indicator, which makes it easier to spot mechanically once you know what to look for.

Choosing an RSI Setting for Your Timeframe

The default 14-period setting is the standard for a reason: it's what most charting platforms ship with, what most traders are watching, and widely-watched levels tend to matter more simply because more traders react to them at the same points.

• Shorter periods (7-9): more sensitive, react faster to recent price changes, generate more signals - and more false ones. Favored by day traders and very short-term swing traders who need faster feedback and are willing to filter more noise.
• Default (14): the balance point, and the one worth knowing even if you eventually customize it, since it's the reference every other setting gets compared against.
• Longer periods (21-25): smoother, slower to react, fewer but generally more reliable signals. Favored by swing and position traders holding for weeks rather than days, who want to filter out short-term chop.

There's no universally "correct" setting - it should match your holding period. Changing it without a specific reason usually just adds noise rather than edge.

Combining RSI With Other Signals

RSI is most useful as one input among several, not a standalone trading system:

RSI + trend structure. Use the price trend (higher highs and higher lows, or the reverse) to decide which RSI framework applies - the trend-following range-shift approach or the classic range-bound overbought/oversold approach.

RSI + MACD. MACD is a trend-following momentum indicator built from moving average crossovers; RSI is a bounded oscillator. When both agree - MACD confirming trend direction while RSI shows a failure swing or divergence at an extreme - the combined signal carries more weight than either alone. See the MACD Indicator Guide for the full mechanics.

RSI + Bollinger Bands. Price tagging the upper Bollinger Band while RSI is not overbought can indicate a genuinely strong trend rather than an exhaustion point; price tagging the band while RSI is extended adds confidence to a mean-reversion setup. See Bollinger Bands Explained.

RSI + volume. A divergence or failure swing accompanied by a volume spike carries more conviction than the same pattern on light volume, since volume shows whether real participation backs the momentum shift.

Common RSI Mistakes

1. Trading overbought/oversold as an automatic signal, regardless of trend. The single most common error - shorting every RSI reading above 70 in a strong uptrend, or buying every reading below 30 in a strong downtrend.
2. Ignoring which regime applies. Range-bound rules (sell 70, buy 30) applied to a trending market, or trend-following rules applied to a genuinely choppy, range-bound stock.
3. Trading divergence alone, with no confirmation. Divergence can persist far longer than expected before price actually turns.
4. Using an RSI setting mismatched to your holding period. A day trader using a 25-period RSI, or a position trader reacting to every whipsaw on a 7-period RSI, is fighting their own tool.
5. Treating RSI as a price target. RSI tells you about momentum, not where the price is headed or by how much - it has no concept of dollar levels at all.

Watching RSI Without Staring at Charts All Day

RSI updates every trading day, and the setups it flags - a failure swing forming, a divergence developing, a cross above 70 or below 30 - don't wait for you to be looking at a chart when they happen. Most traders who use RSI seriously aren't recalculating it by hand; they're relying on an alert to flag the moment worth a closer look, then doing the actual trend and volume analysis themselves.

Frequently Asked Questions

What is the RSI indicator?

The Relative Strength Index (RSI) is a momentum oscillator developed by J. Welles Wilder in 1978 that measures the speed and magnitude of a stock's recent price changes on a scale from 0 to 100. It compares the size of recent gains to the size of recent losses over a set lookback period, most commonly 14 days, to gauge whether a stock's recent momentum has been predominantly upward or downward. RSI does not measure price level or trend direction on its own - it measures how fast and how hard a stock has been moving in one direction relative to the other.

How is RSI calculated?

RSI starts by averaging the size of up days and the size of down days separately over the lookback period (usually 14 periods), producing an Average Gain and an Average Loss. Those two averages form the Relative Strength (RS) ratio: RS = Average Gain divided by Average Loss. That ratio is then scaled into the 0-100 range with the formula RSI = 100 minus (100 divided by (1 plus RS)). After the first calculation, most platforms use Wilder's smoothing method, which weights in each new day's gain or loss rather than fully recalculating the raw average, so RSI updates smoothly rather than jumping around.

What do overbought and oversold mean in RSI?

An RSI reading above 70 is traditionally labeled overbought, meaning recent gains have significantly outpaced recent losses, and an RSI reading below 30 is labeled oversold, meaning the reverse. These are starting points for further analysis, not automatic buy or sell triggers. In a strong, sustained uptrend, RSI can remain above 70 for weeks while the stock keeps climbing - selling purely because RSI crossed 70 has caused traders to exit strong trends far too early.

What is RSI divergence?

RSI divergence happens when a stock's price and its RSI reading move in opposite directions, which can signal that the trend driving the price is losing momentum even though price hasn't confirmed it yet. Bearish divergence occurs when price makes a new high but RSI makes a lower high; bullish divergence occurs when price makes a new low but RSI makes a higher low. Divergence is a warning sign to watch more closely, not a standalone trade signal - it can persist for a long time before price actually reverses.

What is the best RSI setting to use?

The default 14-period setting is the most widely used and the one virtually every charting platform ships with by default, which matters because widely-watched levels tend to actually act as levels other traders react to. Shorter periods (7-9) make RSI more sensitive for short-term traders; longer periods (21-25) smooth it out for swing and position traders. There is no single correct setting - it should match your holding period.

What is the most common RSI mistake traders make?

The most common mistake is treating an overbought reading (above 70) as an automatic sell signal, or an oversold reading (below 30) as an automatic buy signal, regardless of the broader trend. In a genuinely strong uptrend, RSI can stay elevated for extended periods precisely because the stock keeps making new highs. The more reliable approach reads RSI's behavior relative to a level - such as whether it holds above 40 on pullbacks during an uptrend - rather than treating any single crossing as a trigger.

Related Reading

• RSI Divergence Explained - the full breakdown of regular and hidden divergence patterns introduced above
• MACD Indicator Guide - the trend-following momentum indicator that pairs naturally with RSI
• Bollinger Bands Explained - combine volatility bands with RSI to separate strong trends from exhaustion
• Moving Averages Explained - the trend context that decides which RSI framework applies
• How to Set Stock Price Alerts - the mechanics behind the RSI alert covered above