Option buying · backtested on 1-minute option prices
Does buying NIFTY and SENSEX options pay?
Nine ways of buying options, from the plain expiry-day straddle to calendar spreads, tested on two years of real option prices (2024-10-03 to 2026-10-01) after brokerage, taxes and slippage. Losers stay listed next to the rest, so nothing is cherry-picked.
Short answer: no. 0 of 9 strategies passed the bar and 8 of 9 lost money net after costs. The plain at-the-money straddle costs more than the market delivers, and no filter tested changed that.
Data to · page updated
Round 1: is volatility cheap?
- Plain ATM straddleFailedNet P&L−₹1,78,044Trades208Win rate33%Profit factor0.60Year 1−₹85,851Year 2−₹92,193
- Cheap-volatility straddleFailedNet P&L−₹72,996Trades91Win rate30%Profit factor0.55Year 1−₹22,411Year 2−₹50,585
- Cheap-volatility strangleFailedNet P&L−₹21,457Trades91Win rate22%Profit factor0.21Year 1−₹11,237Year 2−₹10,220
- Low-volatility regimeFailedNet P&L−₹49,601Trades253Win rate37%Profit factor0.83Year 1+₹8,521Year 2−₹58,122
- Event-day straddleToo few tradesNet P&L−₹34,854Trades28Win rate32%Profit factor0.57Year 1−₹58,027Year 2+₹23,173
- Straddle with sold wingsFailedNet P&L−₹84,311Trades91Win rate30%Profit factor0.49Year 1−₹31,443Year 2−₹52,868
Round 2: different questions
- Gap-day straddleFailedNet P&L−₹52,500Trades176Win rate38%Profit factor0.81Year 1+₹2,502Year 2−₹55,002
- Calendar spreadFailedNet P&L−₹37,300Trades206Win rate50%Profit factor0.89Year 1+₹5,219Year 2−₹42,519
- Confirm, then buyNo edgeNet P&L+₹215Trades76Win rate29%Profit factor1.00Year 1−₹25,516Year 2+₹25,731
What the tests show
On expiry days the at-the-money straddle costs about 0.52% of the index at 09:20, while the index moves about 0.47% by 15:15. The realized move beat the price paid on only about a third of days, which is why buying it loses before any filter is applied.
Filters for cheap volatility, gap days, events and confirmation did not change that picture. The calendar spread lost the least: the expiring leg it sells earns the decay that option sellers collect, and the long leg is the cost.
The pass bar
A strategy passes only if it makes money in year 2, has a profit factor above 1.3, has at least 40 trades, beats its own baseline per trade, and is profitable in all four cells (NIFTY and SENSEX, year 1 and year 2). Year 1 runs to 30 September 2025; year 2 is the unseen test. Rules, exits and sizes were fixed in code before each run and not tuned afterwards.
How the tests were run
Minute-by-minute replay of expired weekly option contracts. Every fill carries 0.5 points of slippage plus brokerage, taxes and exchange charges. One lot: 65 for NIFTY and 20 for SENSEX, so both years are in today's rupees. Everything is bought and sold on the same day; nothing is held overnight.
Frequently asked questions
Does buying NIFTY or SENSEX options make money?
Not in these tests. 9 option-buying strategies were backtested on two years of 1-minute NIFTY and SENSEX option prices after costs, and none passed MarketCue's bar (profit in year 2, profit factor above 1.3, at least 40 trades, better than its baseline, and profit in all four index and year cells). 8 of 9 lost money net.
Which option-buying strategy worked best?
The closest to break-even were Confirm, then buy (+₹215 over 76 trades), Cheap-volatility strangle (−₹21,457 over 91 trades), Event-day straddle (−₹34,854 over 28 trades). None of them passed the bar, and the best result is within the size of slippage and charges, so it is not evidence of an edge.
Why does buying an at-the-money straddle lose money?
On expiry days the NIFTY and SENSEX at-the-money straddle costs about 0.52% of the index at 09:20, while the index moves about 0.47% by 15:15. The realized move beat the price paid on only about a third of expiry days, so the buyer pays more than the market delivers on average. Published research on index and equity options finds the same.
Do filters for cheap volatility, gaps or events help?
They changed the size of the loss, not its sign. Cheap-volatility filters, a low-volatility regime, gap days, event days, selling far wings and a confirmation entry all lost money or were break-even noise, and a calendar spread lost less than the plain straddle because the expiring leg it sells earns the decay.
How were the option-buying strategies tested?
Minute-by-minute replay of expired weekly option contracts, 2024-10-03 to 2026-10-01. Every fill carries 0.5 points of slippage plus brokerage, taxes and exchange charges, one lot (65 for NIFTY, 20 for SENSEX), everything closed the same day. Rules and exits were fixed before each run and not tuned afterwards. Year 1 runs to 2025-09-30; year 2 is the unseen test.
Is this trading advice?
No. These are backtests on historical data, not live results and not a recommendation. Past performance does not guarantee future returns, and buying options can lose the whole premium paid. MarketCue is not a SEBI-registered investment adviser.
Backtests on historical data, not live results and not a recommendation. Past performance does not guarantee future returns. Options trading can lose the whole premium paid. MarketCue is not a SEBI-registered investment adviser. See the disclaimer.