RSI Mean Reversion
Buys short, sharp dips inside an uptrend, adds as the dip stretches, and sells everything at once when the bounce comes.
- Markets
- Liquid coins, futures
- Chart
- Daily candles
- Direction
- Long only
- Leverage
- 2x, isolated margin
- Typical holding
- A few days
- Live since
- 25 July 2026
- The dip begins: the uptrend is intact and the short-term RSI drops into oversold. The first oversold days are only counted, not bought.
- Scaling in: each further oversold day adds one equal tranche, up to three.
- Average price: every tranche bought lower pulls the average entry down.
- Exit: the short-term RSI is back above neutral, and the whole position is sold at the close.
- Short-term RSI: the shaded band is the oversold zone, the dashed line is neutral.
The idea
In a healthy uptrend, sell-offs rarely last long. A few red days in a row push a coin far below where it traded a week earlier, short-term traders give up, and buyers who missed the trend step back in. The very short-term RSI measures exactly this kind of stretch: it drops to extreme lows after only a handful of falling days.
The strategy does not buy the first sign of weakness. Most small dips end before they become interesting, so the first oversold days are only counted. Only if the selling continues does it start to buy, in equal tranches, so the deeper the dip goes, the lower the average price. When the bounce comes, the whole position is sold in one go.
How it trades
Each dip runs through the same sequence, decided at the daily close:
Healthy trend
A new sequence may only begin while the coin trades above its medium-term average.
Oversold
The short-term RSI falls deep into oversold territory. The count starts.
Let it run
The first oversold days are only counted, not bought. Dips that end quickly are skipped.
Scale in
From then on, every further oversold day adds one equal tranche, up to three in total.
Sell the bounce
Once the short-term RSI is back above neutral, the whole position is sold at the close. If that happens before the first tranche, the sequence ends without a trade.
The next dip starts a new count.
The trend check only decides whether a sequence may start. Once the count runs, further tranches are added even if price slips below the average.
When it sells
Recovery
All tranches are sold together at the first daily close with the short-term RSI above neutral. There is no profit target beyond that; the strategy takes the bounce and steps aside.
No stop-loss
There is no stop. Risk is limited by the trend check at the start, by splitting each position into tranches, and by moderate leverage on isolated margin.
Risk
- Moderate leverage of 2x on futures. Every position uses isolated margin, so a single trade can never draw on the rest of the account.
- Each coin has a fixed slot, and each slot is filled in equal tranches, so a single dip never commits the whole slot at the first price.
- Only a limited number of coins can be held at once, which caps how much is invested during a market-wide sell-off.
- No stop-loss. A dip that turns into a lasting decline is held until it bounces, which can take a while.
Where it does well, and where it struggles
Does well in
- Uptrends with short, sharp pullbacks that are bought back within days.
- Markets where sell-offs overshoot and snap back.
Struggles in
- A trend that breaks: the dip keeps going after all tranches are in, and the position sits at a loss until a bounce comes.
- Long bear markets, where few sequences may start at all.
- Slow, grinding declines without the sharp drops the RSI reacts to.
Backtest
How the strategy would have traded since 2018, with the same rules, limits and leverage as live.
Jan 2018 to Oct 2026. Coins: the 30 largest by market value at the start of each year. Up to four positions of three tranches, 2x leverage. Starting balance $10,000, profits reinvested, fees included.
- Average return per year
- +20%
- Worst drawdown
- −36% −17% on closed trades
- Trades
- 191 About 3 days each, typically
- Winning trades
- 83% Profit factor 3.9
Including open trades at their lowest pointClosed trades only
| Year | Return | Worst drawdown |
|---|---|---|
| 2018 part | −9% | −24% |
| 2019 | +11% | −36% |
| 2020 | +14% | −13% |
| 2021 | +78% | −18% |
| 2022 | +10% | −16% |
| 2023 | 0% | −21% |
| 2024 | +32% | −21% |
| 2025 | +37% | −18% |
| 2026 to Oct | +18% | −10% |
Hypothetical results from a simulation on historical data, not live trading. Live results differ, and past performance is no guarantee of future results. How the backtests were made