How to Trade With RSI(14)

RSI is the most misread indicator on most charts: the 70 and 30 lines that beginners trade as buy and sell buttons are nothing of the kind. This guide covers how RSI(14) is built, why overbought and oversold are myths in a trend, how divergence actually works, and how to use RSI as confirmation behind structure instead of as a standalone signal. Research and education only — not financial advice.

The short answer

RSI(14) is a momentum oscillator: it measures how fast price has risen versus fallen over the last 14 periods and plots the result on a 0–100 scale. The single most important thing to know about it is what it is not — overbought (above 70) is not a sell signal, and oversold (below 30) is not a buy signal. In a strong trend RSI can sit pinned above 70 for weeks while price keeps climbing, and every trader who shorted the first "overbought" reading got run over. Used correctly, RSI is a confirmation and divergence tool that reads the momentum behind a move you already have a structural reason to take — never a trigger on its own.

How RSI(14) is actually built

RSI stands for Relative Strength Index. Over the lookback window — 14 periods by default — you separate the up-closes from the down-closes, average each, and take their ratio:

RS = average gain ÷ average loss | RSI = 100 − [100 ÷ (1 + RS)]

Two facts fall out of that formula and matter more than any rule about it. First, RSI is bounded between 0 and 100 — it cannot run away like price can, which is exactly why it is useful for comparing the strength of one move against another. Second, it is an average of bars that already printed, so like every momentum tool it lags: RSI describes the recent past, it does not forecast the next bar. A reading of 50 is the neutral line where average gains and average losses balance; above 50 momentum leans up, below 50 it leans down.

The overbought and oversold myth

The 70 and 30 lines get taught as reversal zones. They are not. All an RSI of 75 tells you is that recent gains have badly outweighed recent losses — momentum is stretched, not that price must turn. Whether "stretched" means "about to snap back" or "strong and continuing" depends entirely on the regime, and reading the bands without reading the regime is how the myth costs money.

Market regime	RSI above 70 means	RSI below 30 means
Range-bound / mean-reverting	Upper extreme; fades back toward 50 are common	Lower extreme; bounces back toward 50 are common
Strong uptrend	Healthy momentum — RSI can hold above 70 for a long stretch	Rare; dips often floor near 40, not 30
Strong downtrend	Rare; rallies often cap near 60, not 70	Healthy downside momentum — RSI can stay below 30

The practical takeaway: the 70/30 bands are informative in a range and actively misleading in a trend. Some traders shift the bands to 80/20 in trending names; in an uptrend, a pullback that holds around RSI 40 and turns up is often a better tell than anything happening at 70.

Overbought is a description, not an instruction. "RSI is overbought" means momentum is stretched to the upside. Whether that resolves as a reversal or a pause inside a continuation is a question about structure and trend, not about the 70 line. Trading the band alone means fighting strong trends at the worst possible moment.

Divergence: the reading that earns its keep

Where RSI adds genuine information is divergence — when price and momentum disagree. Because RSI is bounded, a second push in price that produces a weaker RSI reading tells you the fuel behind the move is thinning even though price hasn't turned yet.

Regular bearish divergence: price prints a higher high, RSI prints a lower high. The rally is being made on less momentum than the last one — a warning that the uptrend is tiring.
Regular bullish divergence: price prints a lower low, RSI prints a higher low. Selling pressure is fading even as price makes a new low — a warning the downtrend may be stalling.
Hidden divergence: the continuation cousin — price makes a higher low while RSI makes a lower low (bullish), often read as a trend-resumption clue after a pullback.

The unbreakable caveat: divergence is a warning, not a trigger. It is notoriously early, and a strong market can make higher highs on weakening RSI for far longer than an early short-seller can stay solvent. Divergence tells you to pay attention to the level where the trend would break; it does not tell you to enter.

Why RSI alone is not a signal

No oscillator converts to an edge by itself, and we published the evidence against our own signals rather than hide it. In a hypothetical backtest, our raw scanner traded blind — mechanical entries, no catalyst check, no structure, no adversarial review — and produced 161 simulated trades at a 46.6% simulated win rate with a 0.82 simulated profit factor, roughly −2% simulated expectancy per trade. Indicator thresholds like 70/30 are exactly the knob that manufactures a good-looking backtest by luck: tune the level and the lookback across enough history and something will shine by coincidence. The full workings sit in the open on our public record.

Combining RSI with structure — a worked example

Numbers below are illustrative, for teaching the mechanics — not a recommendation.

Context. A stock in a clean uptrend rallies to a new high at $52, having made a prior high at $50. RSI at the $52 high reads 65; at the $50 high it read 74. That is regular bearish divergence — a lower momentum high against a higher price high.
Do not act on the divergence. It is a flag, not an entry. Price is still above support. Shorting here is fighting a live uptrend on a warning.
Wait for structure to break. Mark the level the uptrend must hold — say a prior swing low and rising trendline converging near $49. The trigger is a decisive break and hold below $49 on rising volume, with RSI now losing the 50 midline. Divergence flagged the risk; structure confirms it.
Define the exit before entry. Stop above the $52 high, at $52.20. If short from $48.80, risk per share = $52.20 − $48.80 = $3.40.
Set targets against that risk. TP1 at 1R = $45.40; TP2 at 2R = $42.00 — a defined 1:2 payoff you can check on the risk-reward calculator before committing a cent.
Add a time-stop. If the break below $49 never holds, or the trade stalls for several sessions, the thesis is stale — stand aside. A divergence that never confirms is not a trade.

Notice the rank order: RSI supplied the warning, structure supplied the trigger, and neither alone was the trade. Reverse it — enter on the divergence and hope structure catches up — and you are back to trading the myth.

How our desk uses it

On our model desk, RSI and other momentum tools never occupy the catalyst seat. The full-market scan finds unusual movement; a catalyst check explains it; an adversarial review argues against it; a liquidity screen checks the exit. Only then does a setup become a card carrying a written trigger, TP1/TP2, a stop, and a time-stop, posted before the move to a public, timestamped paper record where the losers stay on the board — the mechanics are on the signals page. RSI's job there is narrow: confirm whether momentum agrees with the direction the catalyst and structure already point to. A bearish catalyst on a name whose RSI is still ramping through 70 is a contradiction worth noticing before entry, not after.

The one-sentence version: RSI answers "is momentum stretched, and does it agree with my structure?" — a useful question — and cannot answer "will price reverse here?" Trading the 70/30 lines as buy and sell buttons treats the first answer as the second, which is why the overbought myth is the most expensive habit RSI teaches.

Common questions

What does RSI(14) actually measure?

RSI(14) is a momentum oscillator that compares the average size of up-closes to the average size of down-closes over the last 14 periods, scaled from 0 to 100. Fifty is the neutral line; above it recent momentum leans up, below it leans down. Because it averages bars that already printed, it describes momentum after the fact rather than predicting the next move.

Is RSI above 70 a sell signal?

No, and treating it as one is the classic mistake. An RSI above 70 only means momentum is stretched to the upside — in a strong uptrend it can stay above 70 for weeks while price keeps rising. The 70/30 bands are most informative in a range-bound market and actively misleading in a trend, where a pullback holding near RSI 40 often matters more than anything at 70.

What is RSI divergence and can I trade it directly?

Divergence is when price and RSI disagree — for example price makes a higher high while RSI makes a lower high (bearish divergence), signaling the move is running on less momentum. It is a warning, not a trigger: markets can make new highs on weakening RSI for a long time. The disciplined use is to let divergence flag the level, then wait for structure to actually break before acting.

Can you trade profitably on RSI alone?

No indicator guarantees profit, and RSI alone is weak — its thresholds whipsaw and its divergences fire early. Our published hypothetical backtest of raw mechanical signals returned a 0.82 simulated profit factor and negative simulated expectancy, which is what indicator-only trading tends to look like. On our desk RSI is a confirmation cross-check behind a catalyst and a structural trigger, never the reason for a trade.

See the process with your own eyes. The desk posts trigger-based cards to a public, timestamped record — losses included — and published the backtest where its own raw scanner loses. The scoreboard is free to watch. Join the floor →

Free to join · paid floors optional · research and education only

Last updated 2026-07-11 · ClaudeQuantAlgo Research Desk · research and education only.

Disclosures. ClaudeQuantAlgo is a research and education community. Nothing on this page constitutes financial, investment, legal, or tax advice, or a recommendation to buy or sell any security or derivative. No profit promises are made, ever — trading stocks, options, and forex involves substantial risk of loss; options positions can lose 100% of their value. The public scoreboard reflects a model ("paper") desk — no real money. Past performance — real, paper, or simulated — never guarantees future results. Hypothetical and simulated results have inherent limitations and no representation is made that any account will or is likely to achieve similar profits or losses. ClaudeQuantAlgo is not a registered investment adviser or broker-dealer. You are solely responsible for your own trading decisions. Never risk money you cannot afford to lose.