RSI 2 Strategy for Indian Markets: Rules and Worked Rupee Examples

What the RSI(2) Strategy Actually Is

The RSI(2) strategy was popularised by Larry Connors and Cesar Alvarez in their book Short Term Trading Strategies That Work. It uses the Relative Strength Index set to a period of 2 instead of the usual 14. A 2-period RSI is extremely sensitive, so it swings between near 0 and near 100 almost daily. That sensitivity is the point: it lets you spot very short bursts of oversold selling inside an asset that is still in an overall uptrend.

A common myth, repeated in many low quality articles, is that you should buy when RSI(2) crosses above 10 from below and short when it crosses below 90. That framing misses the whole idea. The original Connors method is a buy the dip, mean reversion strategy. You are not trying to catch momentum continuation. You are betting that a brief pullback in a strong instrument will snap back. Getting this distinction right is the difference between a tested edge and random clicking.

Because the signal fires often and reverses quickly, RSI(2) is a swing strategy held for one to five trading days, not a scalp and not a long-term investment. It suits the Indian cash segment for delivery trades and index futures for leveraged swings, provided you respect position sizing and the trend filter described below.

The Exact Rules We Will Trade

Here is the clean, testable rule set used for the examples on this page. It is the standard Connors long-only version, adapted for Indian instruments. Long-only matters in India because most retail traders cannot easily short stocks in the cash segment for multiple days, and shorting against a primary uptrend has a poor record.

• Trend filter: only take long trades when the closing price is above the 200-day simple moving average (the 200 DMA). This keeps you on the right side of the long-term trend.
• Entry trigger: buy at or near the close when the 2-period RSI closes below 10. A stricter version waits for RSI(2) below 5 for fewer but higher quality signals.
• Scale-in option: if price falls further the next day and RSI(2) is still below 10, you may add a second equal tranche. This is optional and increases risk.
• Exit trigger: sell when the closing price moves back above the 5-day simple moving average (the 5 DMA). This is the primary exit and replaces a fixed RSI target.
• Hard stop: a protective stop, for example 3 to 5 percent below entry on a stock or a fixed rupee stop on futures, in case the dip turns into a full trend reversal.

Worked Example 1: Reliance in the Cash Segment

These figures are illustrative and rounded for teaching. Suppose Reliance Industries is in a clear uptrend, trading well above its 200 DMA at around 2,850. Over three sharp down days the stock falls to a close of 2,760 and the 2-period RSI closes at 6, a deep oversold reading. The trend filter is satisfied and RSI(2) is below 10, so the rules give a buy signal.

You buy 200 shares at 2,760, a position value of 5,52,000 rupees, taken as delivery. Three trading days later Reliance bounces and closes at 2,830, back above its 5-day moving average, which is your exit signal. You sell all 200 shares at 2,830, an exit value of 5,66,000 rupees. Gross gain is 70 rupees per share times 200 shares, which is 14,000 rupees before costs.

Now the real-world costs on an equity delivery trade. Brokers like Zerodha charge zero brokerage on delivery, so the main charges are STT at 0.1 percent on both buy and sell, exchange transaction charges, GST, SEBI fees and stamp duty. The table below shows the approximate breakdown.

So a clean RSI(2) bounce on Reliance nets roughly 12,756 rupees on 5.52 lakh deployed, about 2.3 percent in three trading days. Held under one year, this profit is short-term capital gains on equity, taxed at 20 percent plus cess. That is roughly 2,551 rupees of tax, leaving about 10,205 rupees in hand. Note that not every signal works, and a single failed trade where the stock keeps falling can wipe out several winners, which is why the stop loss is non-negotiable.

Worked Example 2: Bank Nifty Futures

Index futures let you apply RSI(2) with leverage, but leverage cuts both ways. These numbers are again illustrative. Suppose the Bank Nifty is trending up above its 200 DMA. After a two-day slide the index closes at 48,200 with RSI(2) at 8. The trend filter and the entry trigger both agree, so you buy one lot of the current-month Bank Nifty future.

The Bank Nifty futures lot size is 30. One lot at 48,200 has a contract value of 48,200 times 15, which is 7,23,000 rupees, but you only post margin, roughly 1.1 to 1.3 lakh, so this is a leveraged position. Two days later Bank Nifty reclaims its 5-day moving average and closes at 48,950. You exit one lot at 48,950.

A 750-point capture on one Bank Nifty lot nets about 11,014 rupees. The same move against you, if the index fell 750 points instead, would lose a similar amount, which on 1.2 lakh of margin is close to a 9 percent hit. That asymmetry is why a hard stop and small size matter far more in futures than in cash. Profits and losses from F&O are treated as business income in India, taxed at your applicable slab rate, not at the flat capital gains rate, and they must be reported on the correct ITR form.

Why the 200 DMA Trend Filter Is Non-Negotiable

RSI(2) buys weakness, and weakness in a downtrend is usually the start of more weakness, not a bounce. The 200-day moving average filter is what separates a profitable edge from a falling-knife disaster. In Connors' own published tests, removing the long-term trend filter sharply reduced returns and increased drawdowns. The logic is simple: in an uptrend, dips get bought; in a downtrend, rallies get sold.

On Indian instruments this is especially important during sharp corrections. A stock like a beaten-down PSU or a broken mid cap can show RSI(2) below 5 for days while it keeps grinding lower. The trend filter would have kept you out of that name entirely because the price was below its 200 DMA. Apply the strategy to leaders that are clearly above their 200 DMA, such as a strong large cap or the index itself during a bull phase.

RSI(2) Versus the Classic RSI(14) Approach

Traders often confuse RSI(2) with the textbook RSI(14) overbought and oversold method. They are different tools with different jobs. The table below contrasts them so you do not apply RSI(14) thresholds to an RSI(2) chart by mistake.

Neither is better in the abstract. RSI(2) is a tight, rules-based swing system. RSI(14) is a general-purpose momentum gauge. If you read an RSI(2) chart expecting it to sit politely between 30 and 70, you will be confused, because at period 2 it lives at the extremes by design.

Position Sizing and Risk for Indian Accounts

Mean reversion strategies have a high win rate but occasional large losers when a dip becomes a crash. The only thing that keeps RSI(2) profitable over a year is making each loss small relative to your account. A practical rule is to risk no more than 1 to 2 percent of your capital per trade, measured from entry to your hard stop.

• Decide your stop first, then size the position so the rupee distance to the stop equals 1 to 2 percent of capital.
• In the cash segment, prefer delivery on liquid large caps so you are not forced to square off intraday by margin rules.
• In futures, remember the fixed lot sizes: Nifty 75, Bank Nifty 15, FinNifty 25 and Sensex 10. One Nifty lot at 24,000 is a 18 lakh notional, so even one lot is a large exposure for a small account.
• Never average down past your planned stop. RSI(2) already builds in an optional single scale-in; going beyond that turns a small mistake into an account-ending one.
• Keep a trading journal of every signal, including the ones you skipped, so you can measure your real win rate against the backtest.

Where and When RSI(2) Fails

Honest strategy education means naming the failure modes. RSI(2) breaks down in three situations. First, strong downtrends: the oversold signal fires repeatedly while price keeps falling, so the trend filter is essential. Second, illiquid or news-driven names: a stock gapping down on bad earnings or regulatory action will not mean-revert just because RSI(2) hit 4. Third, choppy, directionless markets, where the 5 DMA exit whipsaws you in and out for tiny losses after costs.

There is also a structural risk specific to Indian retail traders: costs and taxes can quietly eat a thin edge. Each round trip carries STT, exchange charges, GST and stamp duty, and frequent trading in F&O is taxed as business income at slab rates. If your average win is only a few hundred rupees, costs can turn a positive raw strategy into a losing real one. Size up to where the edge is meaningful, or trade less often and only on the cleanest signals.

How to Backtest and Track RSI(2) Honestly

Before risking real money, test the exact rules on at least three to five years of NSE data covering both bull and bear phases. A strategy that only looks good in the 2020 to 2021 bull run is not validated. Track win rate, average win, average loss, maximum drawdown and, critically, net profit after realistic costs and taxes, not the gross point capture.

• Test across regimes: include a falling market like a sharp correction so you see the trend filter actually saving you.
• Record net rupee P&L after STT, brokerage and slippage, not just RSI points.
• Measure maximum drawdown, the worst peak-to-trough fall, so you know the pain you must survive.
• Forward test on a small live size for a month before scaling. Live slippage and emotion never match a clean backtest.

Frequently Asked Questions

Sources and Further Reading

For authoritative data and further reading on this topic, refer to Zerodha Varsity, Investopedia and NSE India. Always confirm current rules, rates and contract specifications on the official source before you trade.