Most traders lose. The numbers are public.
SEBI studies the profit and loss of individual F&O traders in India, and the results are stark. Between FY22 and FY24, 93% of more than one crore individual traders lost money in equity F&O, with aggregate losses above ₹1.8 lakh crore (SEBI, September 2024). In FY25 the share was 91%, and individual traders' net losses widened to ₹1,05,603 crore, up 41% from ₹74,812 crore the year before (SEBI study, July 2025).
Most of those traders were not unlucky. They traded without a tested edge, paid more in charges than they realised, and had no hard limit on what a bad day could cost. All three are fixable — and fixing them is what finding a profitable strategy actually means.
What “profitable” means, in one formula
A strategy is profitable when its expectancy after costs is positive over a large enough sample of trades:
Take a strategy that wins 40% of the time, makes ₹3,000 on an average win and loses ₹1,500 on an average loss. Before costs it earns 0.4 × 3,000 − 0.6 × 1,500 = ₹300 a trade. If brokerage, taxes and slippage cost ₹250 a round trip, the real edge is ₹50 a trade — one bad fill away from zero. That is why a backtest that ignores Indian charges is worse than no backtest: it shows you an edge that does not exist.
- Win rate alone means nothing. A 90% win rate with rare, huge losses is exactly how many option-selling accounts blow up.
- Sample size matters. Twenty trades prove nothing. Judge a strategy on hundreds, across different market conditions.
- Drawdown decides whether you survive to collect the edge. Size so that the worst losing streak in your test would not end your trading.
The six-step process that finds an edge
- Start from an idea you can explain. Why should it work, and who is on the other side? “Breakouts from the opening range on unusually heavy volume tend to continue” is testable. “This stock looks ready to fly” is not.
- Write it down as rules a machine can check. Instrument, timeframe, entry, exit, stop, position size and the hours it may trade. If a rule needs judgement at the moment of entry, you cannot test it — and you will not follow it under pressure either.
- Backtest it honestly. Signals on closed candles only, fills at the next bar’s open, stops assumed to hit before targets inside a bar, every trade charged brokerage, STT, exchange fees, GST and stamp duty, and option legs priced with time decay — never “the index moved 100 points, so the option made 100 points”.
- Try to break it. Test a period you did not tune on, and different market regimes. If the result survives only at one exact parameter value, it is curve-fitted and will not survive live.
- Paper-trade it on live prices. Run the same rules for a few weeks without real orders. Honest paper fills buy the ask and sell the bid; if the edge disappears here, it was never there.
- Go live small, with hard limits — then review every trade. Fix a daily loss cap, a maximum number of trades and a stop on every position before the first live order. Compare each trade with what the plan said to do, and let the numbers, not your mood, decide when to scale up.
Strategy families Indian traders test
None of these is profitable by default. Each has made and lost fortunes; the edge, if there is one, is in the specific rules you test and the discipline you run them with.
| Family | The idea | What usually kills it |
|---|---|---|
| Breakouts & trend following | Ride a move once price leaves a range, such as the opening range, with volume behind it | False breakouts, slippage at the open, choppy days |
| Mean reversion | Fade stretched moves back towards a reference such as VWAP | Trend days, gaps, averaging into a falling position |
| Option selling | Collect time decay with straddles, strangles or defined-risk iron condors | Sudden large moves, event days, margin calls, undefined risk |
| Option buying | Leverage a fast directional move | Time decay and falling implied volatility when the move is late or small |
| Swing & positional | Hold multi-day trends, pullbacks or breakouts in liquid stocks | Overnight gaps, news, holding losers too long |
Red flags when someone sells you a strategy
- Guaranteed or fixed returns. No honest person can promise them in markets.
- A win rate without the average win, average loss and costs. It hides everything that matters.
- P&L screenshots instead of a verifiable record. A screenshot shows the winners someone chose to show you.
- Tips on Telegram or WhatsApp from people who are not registered. In India, only SEBI-registered investment advisers and research analysts may give buy or sell recommendations. Check the name on SEBI’s register of intermediaries before you pay anyone.
How QuantumCat runs the whole loop
QuantumCat is a desktop trading terminal for Indian markets that runs on your own Zerodha, Dhan, Angel One or Groww account. It is built around exactly this process, so every step has a tool — and they share one engine, so a rule means the same thing in the backtest, the paper run and the live order.
| Step | In QuantumCat |
|---|---|
| Rules | A bot is a strategy document — universe, timeframe, entries, exits, risk — built in the five-step guided builder from indicators, price fields, option-chain metrics, Pulse states and time windows. JARVIS, the built-in AI copilot, can build one for you from a plain-English description. |
| Backtest | A backtester built to disappoint you honestly: closed candles, next-bar fills, stops first inside every bar, estimated Indian charges on every trade, Black-76 option pricing — and it refuses to run a strategy on data it cannot reproduce. |
| Rehearse | Your own tick archive from the day you install, and market replay of any recorded day at 1× to 300×. |
| Paper | Bots start in paper mode, and paper fills cross the spread honestly. The per-bot X-ray shows every condition’s live value and why the bot did or didn’t trade. |
| Live | Risk-sized one-keystroke orders, broker-side stops where your broker supports them, account-wide daily loss and order caps, and a kill switch that confirms flat. |
| Review | The trade X-ray measures win rate, net and expectancy by hour, weekday and instrument from your own fills, and the journal charges what your contract note charges. |
QuantumCat does not sell signals or tips, and it never tells you what to buy. It gives you the instruments to find out — quickly and honestly — whether your idea has an edge. Download QuantumCat free and put your first idea through the loop.
Questions
What is the most profitable trading strategy in India?
There isn’t one that works for everyone, and SEBI’s data shows why chasing one fails: 91% of individual F&O traders lost money in FY25. The profitable strategy is the one you have proven has a positive expectancy after costs — over a large backtest sample, then on live paper trading, then on small live size. QuantumCat is built to run exactly that test on your own broker account.
Can I earn a fixed monthly income from trading?
No one can promise a fixed income from trading, and anyone who does is a red flag. Results come in uneven streaks, including losing months. Treat trading as a business: size every position so a bad month is survivable, cap your daily loss, and judge results over hundreds of trades.
How much money do I need to start trading?
Start with none: paper-trade until your rules show an edge on live prices. When you go live, use an amount you can afford to lose entirely, and make sure charges are a small share of each trade — tiny positions are easily eaten by fixed costs. QuantumCat’s bots run in paper mode by default for exactly this reason.
Is backtesting reliable?
Only when it is honest. Most backtests flatter you by filling at the signal candle’s close, ignoring charges and letting targets hit before stops inside a bar. QuantumCat’s backtester evaluates on closed candles, fills at the next bar’s open, resolves stops first, charges every trade and refuses to run on data it cannot reproduce. Treat any result as directional, then confirm it with paper trading.
Does QuantumCat give trading tips or buy/sell signals?
No. QuantumCat is software, not a SEBI-registered adviser, and nothing in it tells you to buy or sell. It gives you the tools to build, test and run your own rules — and evidence, such as Pulse’s order-flow read, with the reasoning shown.
Education, not investment advice. QuantumCat is software, not a SEBI-registered investment adviser or research analyst, and nothing on this page is a recommendation to buy or sell any security. Figures from SEBI and the brokers are linked to their sources; broker prices and plans change, so check each broker's own page.