ALGOINDICATEUR
Login Create an account
🤖 Algo AI Trader 🧠 AI Assistant 📡 Algo Signals VIP 💹 Trading

Platform

📈 Markets ⭐ Watchlist 🌍 Market movers

Tools

🧮 Trader calculators 🔄 Converter 📅 Economic calendar 📖 Trading glossary 🎓 Trader academy 📓 Trade journal 🧰 Advanced calculators 📊 Chart patterns ✨ App features

Resources

📰 News ❓ FAQs ℹ️ About 📨 Telegram signals
Create an account Login Become VIP
Trader academy

The short RSI: buying a pullback within a trend

How the 2-candle RSI, filtered by a trend average, looks for a return to the mean, and why it may buy a fall that keeps going.

Lesson 42 / 96 · 7 min read · Intermediate

Some traders believe that after a very sharp pullback, a trending market tends to return toward its average. This is the idea of “mean reversion”. Algo Signals’ short RSI turns it into a rule: an extremely short RSI, which swings fast, combined with a trend average that says which way to play. It is an idea, not a law.

How it is calculated

It is the same calculation as the RSI (Wilder’s average of closing gains and losses, RSI = 100 − 100 ÷ (1 + RS)), but over only 2 candles. With such a short period the RSI jumps from 90 to 5 in a candle or two. In addition, Algo Signals computes a simple moving average of closes over 200 candles, the “trend average”.

  • RSI period: 2 by default. It reacts almost every candle.
  • Trend average: simple average of the last 200 closes.
  • Default levels: 10 (oversold) and 90 (overbought), far more extreme than 30 and 70.
Real case — ETHUSDT, M15, 2026-10-17 21:15 UTC: the indicator on the chart. Past example, not a forecast.

Reading the signal in Algo Signals

Buy when the short RSI is below 10 AND the close is above the trend average: you buy a pullback within an uptrend. Sell when it is above 90 AND the close is below the average: you sell a bounce within a downtrend. In every other case the state is neutral: the signal lasts only while both conditions are met; it is not held.

Real case — BTCUSDT, H1, 2026-05-01 20:00 UTC: winning signal, target reached. Past example, not a forecast.

The settings

  • Period (period): 2 by default (1 to 20). Longer, the RSI becomes slower and closer to the classic RSI.
  • Oversold (r2Lo): 10 by default (1 to 49). Overbought (r2Hi): 90 by default (51 to 99). More extreme, fewer signals; closer, more of them but more fragile.
  • Trend average period (trend): 200 by default (20 to 1000). Shorter, it lets more signals through and filters the trend less well.
  • These are starting values, not proven optima.

The pitfalls

This RSI buys a fall: when it is wrong, it is wrong in the middle of a decline. A very low RSI can stay low as long as the market keeps collapsing, and bad news or a crash can break the trend average itself. The “above the average” condition is only a filter, not insurance. It also says nothing about the destination: the rebound may be tiny. On spike indices (Boom/Crash), sudden spikes often trigger it without the “mean reversion” logic applying.

Real case — BTCUSDT, H1, 2026-09-04 02:00 UTC: losing signal, stop hit. Past example, not a forecast.

Key takeaways

  • 2-candle RSI, levels 10 / 90, filtered by a 200-candle average: you buy a pullback in an uptrend and sell a bounce in a downtrend.
  • The signal is not held: it disappears as soon as the conditions are no longer met.
  • It buys a fall: a stop is essential and position size should stay small.
  • Combine it with a trend-strength indicator (ADX) or a support zone to avoid falls that keep going.

📝 Lesson test

Log in to take this lesson’s test. Once you have passed all the tests, you will receive a certificate in your name.

Log in

These lessons are for understanding and are not investment advice. No method guarantees winning.