RSI — foundations

Core idea

RSI isn't "a momentum indicator for catching overbought/oversold," it's a ratio between buying pressure and selling pressure present in the market on a given timeframe. Being able to read the buy/sell relationship already answers the root of the question "will price go up or down." In the teacher's system, RSI is the only indicator used — everything else just adds noise.

Details

What it really is: a fraction

RSI ≈ (buying pressure) / (selling pressure), scaled to 0–100. A metaphor the teacher likes to use: a rice market with 100 people in it. 60 buyers / 40 sellers → price rises; 40 buyers / 60 sellers → price falls; 50/50 → price stands still. RSI is literally that percentage.

Because it's a fraction, RSI rising has two possible causes, and RSI falling also has two:

How to tell the two cases apart — this is exactly where the index zone helps:

• RSI rises but only flounders in the 30–40 zone → most likely it's just selling pressure weakening, no real buying pressure yet.
• RSI rises and climbs above 60 → confirms buying pressure has genuinely increased.
• RSI falls but still holds above 40 → it's just buying pressure resting, selling pressure hasn't come in.
• RSI falls and breaks below 40 → selling pressure has genuinely increased, it's no longer simply "the buyers went on a break."

This is one of the most valuable parts of reading RSI this way: it tells you who is creating the move, not just which direction the move is in.

Setup

• RSI period stays at the default 14, never changed.
• Turn everything else off to keep it clean: turn off the default "RSI-based MA" line, turn off the overbought/oversold shading.
• Then apply 2 moving averages onto RSI itself (EMA 9 and WMA 45) — see rsi trend and the two moving averages.
• Do not apply a moving average onto price.

The index zones and their meaning (memorize this)

A direct consequence: if an entire stretch of RSI sits squarely inside 40–60, you can tell price is sideways on that timeframe without even looking at the price chart.

In theory RSI runs 0–100, but in practice it never touches either end, because the market is never absolutely one-sided. In practice it only wanders around 10–90.

Cyclicality (a sine-wave shape)

Price has no bound (30k today, 200k next year), but once you convert it to a buy/sell ratio, RSI always alternates from its lower edge to its upper edge and back, because crowd psychology alternates between greed and fear. Sell too much and it creates a need to buy; buy too much and it creates a need to take profit.

• During a strong uptrend, that sine-wave swing skews heavily toward the upper half of the scale.
• During a downtrend it skews toward the lower half.
• During a range, it oscillates narrowly around the middle, and you have to look at the whole shape, not get fooled by the small, fidgety crosses up and down.

Putting it into practice

1. Look at RSI to determine the buying/selling pressure relationship on each timeframe — not to catch a direct buy/sell signal.
2. Pair the index zone with RSI's direction to answer: is this stretch of price being driven by one side, or is it just one side taking a rest.
3. RSI sitting squarely in 40–60 → that timeframe has no opinion, don't trade off that timeframe.
4. RSI signals extremely strong buying pressure (>80) → your job is to go with it, wait for a pullback to buy, don't go hunting for a top. The opposite applies below 20.
5. The system's full toolkit: RSI accounts for ~80% of the decision, the remaining 20% is support/resistance + basic Dow theory + basic Fibonacci.

What the teacher stresses / common traps

Drop the term "overbought / oversold" entirely. The teacher hates these two words because they plant a reversal suggestion in your head: see RSI > 70 and go looking for a short, see < 30 and go try to catch the bottom — that is, you put yourself against the whole market. Replace it with neutral language: "buying pressure is currently well above selling pressure." An example the teacher gives: BTC's weekly RSI broke above 80 in November 2020 when price was still around 17,500 and stayed above 80 all the way up to 65,000 — anyone who saw "overbought" and shorted blew up however many accounts they had. Even when RSI had only just touched 70, price was only 10,000–12,000.

The argument by money: RSI at 80 means 80% of the market is buying. To push price the other way, you'd have to put up more money than the total buying pressure of that 80% just to absorb it all before you could even start pushing price down. Nobody can do that → don't fight it.

The secretary metaphor: RSI is the secretary who runs in to report the news. If she says "it's 40 degrees and sunny outside," you carry an umbrella — you don't infer "it's this sunny, rain must be coming" and put on a raincoat instead. If she says "it's pouring rain," you put on a raincoat — you don't think "heavy rain means it's about to clear up." This is the concrete version of "Trade what you see, not what you think."

The one exception: the mindset of "sell the high zone, buy the low zone" only works inside a sideways market, and it still has to be combined with other knowledge — see range vs. trend tactics.

Divergence: do not use it. In this group, absolutely do not take a divergence signal (price and RSI out of phase) as grounds for entering a trade. Two reasons:

• Divergence does not signal a trend reversal. It only says the force (buying or selling) is gradually weakening — it could easily just be a correction leg, after which the old trend continues.
• Divergence is a retrospective signal: price has to move first before you can see the divergence. Trading off it means entering late and it's easy for it to turn into guessing. At most, use it as a supporting fact, and only during a range, not during a trending phase.

Don't stack on more indicators. No MACD, no Stochastic, no Ichimoku, no extra lines on price. The teacher's reasoning: every trading system in existence is built on the same foundation, the price chart; dig deep enough and they all lead to the same conclusion, and adding indicators only clutters the decision-making process. The classic trap is "carving the plow in the middle of the road": the system runs well in a trend, gets noisy in a range, people think the system is wrong and go learn a new one — keep repeating that and you never get good at anything.

A trap about the level of understanding: understanding it while listening to the lecture ≠ being able to analyze it yourself. The teacher states the timeline plainly: it's around session 20–25 before you start keeping up with the analysis, around session 30 before you can work out the logic yourself. Before that, absolutely do not trade with real money.

Related lessons