Methodology
How the numbers are calculated
Every figure on the strategy pages comes from a back-test run on historical market data — not a projection and not a promise. This page states exactly how each number is produced and where its limits are, so you can judge it for yourself.
How every strategy is tested
01
Historical data
Minute and daily candles for NIFTY, BANKNIFTY and SENSEX and their option chains, sourced from the exchange feed and broker historical APIs. Index prices are the same single source that drives the live engine.
02
Rules, fixed in advance
Every entry, exit, stop-loss, target and square-off is a mechanical rule defined before the test — no discretion, no hindsight. The same rule code runs the backtest and the live strategy.
03
One lot, honest fills
Each trade is simulated at one lot using the exchange lot size for that period. Premiums come from the recorded chain, not a model. Results are stated per lot so you can scale to your own size.
04
Costs stated separately
Headline figures are BEFORE brokerage, slippage and taxes so the strategy edge is visible on its own. Your net will be lower — size and cost assumptions are yours to apply.
05
Full window, no cherry-picking
Metrics span the entire available window (e.g. 2021–26; SENSEX from 2023). We do not trim losing months or start dates to flatter a curve.
What each metric means
Win rate
Share of closed trades that finished profitable, over the full test window. A high win rate does not guarantee a positive edge on its own — read it with profit factor.
Profit factor
Gross profit ÷ gross loss. Above 1.0 means winners outweigh losers; the selling strategies here sit around 6–7, meaning ₹6–7 earned per ₹1 lost across the sample.
Max drawdown
The largest peak-to-trough fall in cumulative P&L, per lot. This is the loss you must be able to sit through — not the worst single trade.
Annualised return (CAGR)
Computed from total P&L over the margin/capital base across the actual number of years tested — never a headline figure typed by hand.
Total P&L
Cumulative profit for ONE lot over the whole window, before brokerage, slippage and taxes. Multiply by your lots, then subtract costs, for a real-world estimate.
Why the library is option-selling only
We back-tested buying strategies too — intraday breakouts and momentum-buying among them. On the same historical data and the same honest rules, they lost money across the window. Rather than dress a losing edge in a nice chart, we excluded them. The strategies you see are the ones that actually held a positive edge in the test — all on the selling side, where time decay works for the position. Selling naked also carries larger tail risk, which is why the builder lets you add a hedge leg to cap it.
The honest limits
- !A back-test is not live trading. Real fills, liquidity, gaps, broker latency and your own timing will differ from the simulation.
- !Figures are before brokerage, slippage and taxes. These reduce every result — sometimes materially on high-frequency rules.
- !Past performance is not indicative of future results. A regime the strategy never saw in the window can behave differently.
- !Numbers are per single lot. Scaling up changes fills and market impact; size responsibly.
Rapfin Autopilot is trading-automation software. Rapfin is not SEBI registered and does not provide investment advice, recommendations or tips. You run your own strategies or templates on your own broker account, with your own funds. All performance shown across the site is simulated / back-tested and is not indicative of future results. Nothing here is a solicitation to buy or sell any security.