RSI 2 pullback
The 2-period RSI pullback is a well-known short-term mean-reversion system. A two-day RSI swings to extremes quickly, so a reading below 5 marks a sharp short-term sell-off and above 95 a sharp spike. This version buys NIFTY futures after deeply oversold daily closes and shorts after deeply overbought ones, adds on further extremes, and exits once RSI returns to the middle.How it trades
Results
PRE-BUILT · DEFAULTSWhere it made and lost money
With 14 trades, no condition in the evidence reaches the eight trades needed for a firm reading. The splits that stand out are small samples. The four trades entered after choppy, wide-range days lost ₹27,638 together, because a stretched close on such a day was often followed by more of the same rather than a snap-back. The five entries on Tuesday expiry days made ₹48,854, while the other nine lost ₹34,282 overall. Holds through quiet stretches within ±1% lost ₹11,189 over nine trades: when the market drifts, RSI takes days to return to 50 while added lots sit in a losing position.
About this strategy
- After each daily close, compute the 2-period RSI of the NIFTY near-month future.
- If RSI is below 5, buy one lot at the next session's first tick; if above 95, sell one lot.
- While in the trade, each further close beyond the entry threshold adds one more lot, up to three lots.
- Exit the whole position at the next session's first tick after RSI closes back at or beyond 50; the same tick can open a trade the other way.
- There is no stop or target; on the futures expiry day the stack is closed at 15:15 and reopened in the next month's contract.
- An optional filter only buys above the 200-day average and only shorts below it.
Why traders use it
Indices tend to snap back after sharp two- or three-day moves, and a 2-period RSI is a very sensitive way to spot those stretches. Exiting at RSI 50 takes the snap-back and nothing more, so trades usually last only a few days. Adding on further extremes improves the average price if the pullback deepens. The rules are short, fully mechanical and use only daily closes, which makes the system popular with part-time traders.
The year in plain words
On NIFTY futures from 1 Oct 2025 to 30 Sep 2026 the system made 14 trades, won eight and made ₹14,572 before costs, a profit factor of 1.16. After ₹20 per order and bid-ask slippage it was a ₹666 loss, roughly break-even. Average wins of about ₹13,220 were smaller than average losses of about ₹15,200, because losers often carried added lots. Thirteen trades ended when RSI returned to 50 and one at a monthly roll. July 2026 made ₹37,225 and May ₹29,646, while September lost ₹20,709 and March ₹19,188. The maximum drawdown before costs was about ₹51,700.
What the settings do
Risks
- There is no stop loss, and adding lots into a move that keeps going can triple the loss.
- Positions are held overnight and are exposed to gaps; the worst trade lost about ₹20,700.
- The edge before costs was thin enough that costs removed it entirely.
- With about 14 trades a year, a handful of outcomes decide the result.
Who it suits
Short-term swing traders who prefer buying weakness and selling strength on daily data, have margin for up to three NIFTY futures lots and can hold through drawdowns without a stop.
Questions traders ask
What is the RSI 2 strategy?
It uses a two-period RSI on daily closes to find very short-term extremes. Readings below 5 are treated as oversold and bought; readings above 95 are treated as overbought and sold. Positions are closed when RSI returns to around 50.
Is the RSI 2 pullback strategy profitable on NIFTY?
In this backtest from 1 Oct 2025 to 30 Sep 2026 it made ₹14,572 before costs but lost ₹666 after ₹20 per order and bid-ask slippage, over 14 trades. Statutory charges were excluded. Past results do not guarantee future ones.
Why add to a losing position?
The idea is that a deeper extreme means a stronger snap-back, and adding improves the average entry. It also raises risk: if the move continues, the loss grows with each added lot, which is why the number of lots is capped at three.
Does the 200-day trend filter help?
It is off in the default run. Turning it on only buys dips in an uptrend and only shorts rallies in a downtrend, which is a common way to avoid fighting the main trend. Rerunning with it on shows the effect on this period.