Module 4 · Indicators, by family

RSI

Lesson 4.5 · ~8 min read · 24th of ~51

RSI might be the most misused indicator ever invented. Almost everyone learns it the same way — "above 70, sell; below 30, buy" — and almost everyone who actually trades it that way loses money. Not because the indicator is broken, but because that rule is a myth.

RSI is genuinely useful once you understand what it measures and how professionals actually read it. So we're going to unlearn the myth first, then rebuild it into the two uses that hold up: reading momentum with the trend, and spotting divergence.

The idea, in plain language
What RSI actually measures

RSI stands for Relative Strength Index, and it's an oscillator — an indicator that moves within a fixed range, in this case from 0 to 100, drawn in its own little panel beneath the price. What it measures is momentum: the speed and size of recent price moves. Under the hood it compares the average size of the up-moves to the average size of the down-moves over a lookback period (the default is 14 candles) and boils that into a single number.

Read it as a tug-of-war meter. A high RSI means recent candles have been dominated by strong buying — up-moves have been big and frequent. A low RSI means selling has dominated. The 50 mark is the neutral middle: above 50, buyers have had the edge lately; below 50, sellers have. That's the honest core of RSI — it's a momentum gauge, nothing more mystical than that. And because it's built from recent price, it's still, like everything last two lessons, derived from data you already have.

The overbought / oversold myth

Here's where it goes wrong for almost everyone. Two levels get special names: above 70 is called "overbought," and below 30 is called "oversold." The seductive, everywhere-taught interpretation is: overbought means "too high, about to fall — sell," and oversold means "too low, about to bounce — buy." It sounds logical. It is a great way to lose money.

Why? Because overbought means strong, not doomed. A high RSI is telling you buyers are powerful right now — and powerful trends stay overbought for a long, long time while price keeps climbing. Betting against an overbought reading in a strong uptrend is just the same mistake from the moving-averages lesson in a new costume: fighting the current because it "looks too high." The trader shorting every 70 reading in a raging bull move gets run over again and again. Oversold in a strong downtrend is the mirror trap — "it's so cheap it must bounce," right before it gets cheaper.

The myth isn't entirely useless — overbought/oversold reversals do work more often inside a range, where price is bouncing between a floor and ceiling with no trend. But treating 70 and 30 as automatic sell and buy buttons, regardless of context, is the single most expensive habit RSI has taught a generation of beginners. Delete it.

How it's really used

Strip away the myth and two solid uses remain. The first is momentum in agreement with the trend. Instead of fading extremes, you use RSI to time entries in the trend's direction. In an uptrend, RSI pulling back toward the 40–50 zone (not necessarily all the way to 30) often marks a healthy dip — a spot to join the trend as momentum resets, exactly when price is also pulling back to a moving average. The 50 line doubles as a bull/bear filter: momentum holding above 50 supports staying long; breaking below 50 warns the balance is shifting.

The second, and most respected, use is divergence. Divergence is when price and RSI disagree. If price grinds to a higher high but RSI makes a lower high, the new price peak was reached on weaker momentum than the last one — buyers are getting tired even as price rises. That's bearish divergence, a warning the uptrend may be running out of fuel. The reverse — price making a lower low while RSI makes a higher low — is bullish divergence, hinting a downtrend is losing steam. Divergence is powerful because it catches the weakening under the surface before the price structure itself breaks. But hold that power loosely: divergence is a warning, not a trigger, and we'll be honest about its limits below.

See it on a chart

First, the myth in action. Watch price rip higher in a strong uptrend while RSI sits pinned in "overbought" the whole way — and imagine how every naive "sell the 70" trade would have gone:

"overbought" ≠ "about to fall" · RSI can stay hot for the whole trend

price · strong uptrend 70 30 RSI stays overbought — every "sell" here was wrong
↳ RSI holds above 70 for the entire climb, because a strong trend simply is persistent momentum. "Overbought" was describing strength, not predicting a fall. Fade it and you're shorting a freight train. In a trend, high RSI is a reason to stay long, not to bail.

Now the real signal: bearish divergence. Price pushes to a higher high, but RSI quietly makes a lower high — momentum fading beneath a rising price:

bearish divergence · higher high in price, lower high in RSI

price: higher high ↗ 70 30 RSI: lower high ↘ → momentum fading
↳ Price made a new high, but RSI didn't — the second peak came on weaker momentum. That gap is bearish divergence: the crowd behind the rally is thinning even as price climbs. It's an early warning the trend is tiring — but notice it doesn't tell you when the turn comes, only that fuel is running low.
The honest truth

Divergence is real, but it's chronically early and it fails often. A strong trend can flash divergence for weeks, making lower-and-lower momentum highs while price keeps grinding up, and traders who short the first divergence get stopped out over and over before the top finally arrives. Divergence tells you the trend is weakening, never when it will actually turn — so it's a reason to tighten stops or wait for the structure to break, not a standalone sell signal.

And the deeper caveat: RSI is a momentum lens on price, not a separate oracle. It's a single number that throws away most of what the chart is showing, so it's only as good as the context you read it in. Overbought/oversold behaves one way in a range and the opposite way in a trend; the 40–50 pullback zone only means "buy the dip" if there's an uptrend to dip within. Never trade RSI alone. Its job is to add a momentum read on top of the trend and levels you've already established — one voice in the choir from Module 5, never a soloist.

So carry RSI as a momentum companion, tuned to the regime. In a trend: high RSI confirms strength, pullbacks toward 50 time your entries, and divergence warns you the fuel is low. In a range: the 70/30 extremes regain some of their old meaning as bounce zones. The skill isn't memorizing "70 sell, 30 buy" — it's knowing which of those two worlds you're standing in, and letting the same indicator mean different things in each. Next we meet a close cousin, the Stochastic, that measures momentum a bit differently and shines in exactly the range conditions where RSI's extremes come back to life.

Try it yourself

Open the Lab and add RSI below the price. First, find a strong uptrend and just watch RSI camp above 70 while price keeps rising — feel in your gut why "sell overbought" would have been a disaster. Then flip to a clearly range-bound chart and notice how the 70 and 30 touches line up much better with the turns. Same indicator, opposite behavior — that's the whole lesson.

Next, go hunting for a divergence: a spot where price makes a new high (or low) but RSI doesn't. Mark it, then press Play and see what follows — sometimes a clean turn, sometimes the trend shrugs it off and keeps going. Watching divergence both work and fail is what teaches you to treat it as a warning, not a trigger.

Open the Lab →
Three things to keep