Backtesting

Intraday vs positional backtesting: why 1-minute data matters

How intraday and positional option strategies differ in a backtest, why stops need 1-minute candles, and how gaps, expiry and settlement change the results.

By
Veloce team
Published
Reading time
9 min read
On this page
  1. Two strategies, not two settings
  2. How Veloce builds candles from one-minute data
  3. The same stop on one-minute and five-minute candles
  4. Why entry times need one-minute data too
  5. What changes for positional strategies
  6. Gaps
  7. Expiry and settlement
  8. Strike reach on later expiries
  9. A positional spec
  10. The MCX evening session
  11. Choosing a candle size for the test
  12. Sizing and costs across both styles
  13. Try it in Veloce

An intraday strategy closes everything before the session ends; a positional one holds across sessions, so its backtest also has to handle overnight gaps, weekends, expiry and settlement. Both need one-minute data underneath: stops, fills and entry times are only as precise as the candles you replay, a five-minute candle can hide a spike that would have stopped you out, and with one-minute data you can still build any larger candle. This article shows, with numbers, how the same stop behaves on one-minute and five-minute candles, what changes when a position is held overnight, and how to set up both kinds of test in Veloce.

Two strategies, not two settings

It is tempting to treat “square off at 15:15” as a small option on an otherwise identical strategy. In a backtest it changes almost everything:

Intraday Positional
Day starts Flat With yesterday’s positions
Overnight gap Not held Hits open positions at the next open
Weekend and holidays Not held Time decay and news pile up between candles
Expiry Usually not reached, unless trading on expiry day A planned event: exit before it or settle at it
Stops Checked all day, every candle Can only act once the next session opens
Capital at risk Returns to zero every evening Stays open overnight
Typical candle 1 to 5 minutes 5 to 60 minutes, still built from 1-minute data

In Veloce, an intraday strategy has a square_off time; a positional one leaves it empty and uses exit rules such as before_expiry instead. Both run on the same engine, which replays one-minute data and builds the strategy’s candles from it.

How Veloce builds candles from one-minute data

Veloce stores one-minute candles for each index and option contract and builds the strategy’s candle size from them, anchored at the session open: 09:15 for NSE and BSE, 09:00 for MCX. A 5-minute NIFTY strategy sees candles starting 09:15, 09:20, 09:25, exactly as a chart would. The open is the first minute’s open, the close the last minute’s close, the high and low the extremes of the minutes inside.

That has two consequences:

  1. Any candle size is available. A sweep can compare 1, 3, 5 and 15 minute candles on identical underlying data, because they are all built from the same minutes.
  2. The decision grid follows the candle. Exits are checked on each close of the strategy’s candle, and orders fill at the open of the next one. On 5-minute candles, that is every five minutes.

The same stop on one-minute and five-minute candles

Take the sold 25,900 NIFTY call from the short straddle guide: sold at ₹128.35, with a stop at 30% of its credit that triggers on a close of ₹166.10 or more. Here is an illustrative stretch of one-minute candles around a sharp spike:

One-minute candle Open High Close
11:36 ₹146.00 ₹151.40 ₹150.20
11:37 ₹150.20 ₹158.80 ₹157.90
11:38 ₹157.90 ₹168.30 ₹166.80
11:39 ₹167.40 ₹167.90 ₹161.00
11:40 ₹161.00 ₹161.50 ₹154.30

The five-minute candle starting 11:35 is built from the 11:35 to 11:39 minutes, and the next one from 11:40. Run the same strategy on both:

stop: close of ₹166.10 or more five-minute candle 11:35 to 11:40, close ₹161.00 11:38 close ₹166.80: stop on 1-minute 11:3611:3711:3911:40
On one-minute candles the 11:38 close breaches the stop. The five-minute candle containing that minute closes at ₹161.00, below the stop, so a five-minute test does not stop out there.
One-minute candles Five-minute candles
First close at or above ₹166.10 11:38 candle, ₹166.80 None in the 11:35 candle (close ₹161.00)
Stop decision At 11:39 Not triggered
Fill 11:39 open ₹167.40 + 1 tick = ₹167.45 Position stays open
Loss on the leg ₹2,541.50 plus charges Depends on what happens next

Neither result is “wrong”. They are two different strategies: one watches its stop every minute, the other every five minutes. What goes wrong is testing one and trading the other. If you will run your stop on a broker’s order book, which reacts to individual trades, the one-minute test is the closer of the two, and still a little optimistic or pessimistic depending on the path inside each minute.

Two details make the gap larger than it looks:

  • The candle’s high is not a trigger. Veloce checks exits on candle closes, not highs, because a close is a price the position could actually be marked at. The 11:38 high of ₹168.30 does not trigger the stop on its own. Real stop orders trigger on trades, so a live stop could fire on a high that a closing-price test never sees.
  • The fill is the next open, not the stop level. In the one-minute test the stop decided at ₹166.80 and filled at ₹167.45. On five-minute candles, a stop that does fire later fills at the open of the next five-minute candle, which can be minutes and many rupees further away.

Why entry times need one-minute data too

Popular intraday rules are written in minutes: the 9:20 straddle, a 15-minute opening range, an entry at 10:15. An entry time means “the first candle that closes at or after this time”, so it only lands where you meant it when the time falls on the candle grid:

Entry time Candle First close at or after it Fill at the open of
09:20 1 minute 09:20 (the 09:19 candle) 09:20
09:20 5 minutes 09:20 (the 09:15 candle) 09:20
09:20 15 minutes 09:30 (the 09:15 candle) 09:30
09:22 5 minutes 09:25 (the 09:20 candle) 09:25

On a 15-minute strategy, “enter at 9:20” really enters at 9:30, ten minutes later than the rule says. On one-minute data every minute is on the grid, which is one more reason to keep minute-level rules on minute candles.

The same goes for exits at 15:15. Since August 3, 2026, NSE’s F&O session runs to 15:40 alongside a closing auction in the cash market, and in Veloce’s one-minute NIFTY data the index candles go flat from about 15:15 and then jump to the official close. Squaring off at 15:15, and keeping later index candles out of signals, keeps an intraday test on normal trading.

What changes for positional strategies

Gaps

There are no candles between 15:30 (or 15:40) and 09:15. A positional strategy’s stop can only act on the close of the first candle of the next session and fill at the open after that. If NIFTY opens 300 points lower, a short put’s loss is whatever that open implies, not the stop level. A backtest that fills positional stops at the stop price is assuming the market waits.

Expiry and settlement

NIFTY weekly options have expired on Tuesdays since September 2, 2025, BANK NIFTY monthlies on the last Tuesday of the month, and SENSEX weeklies on Thursdays. A positional option strategy either exits before expiry or is settled at it. Veloce does both correctly:

  • A before_expiry exit closes the position on the first candle of a day with that many calendar days or fewer to expiry. With days: 1 and a Tuesday expiry, that is Monday morning. Holidays move the expiry earlier, and the exit with it.
  • Anything still open at the close of the derivatives session on expiry day settles at its intrinsic value against the index close. Veloce books settlement without charges, so add the exercise STT yourself for long in-the-money options held to expiry.

Strike reach on later expiries

Veloce fetches option history for strikes within ten strikes of the money on the nearest expiry, and within three on later expiries. A positional strategy that trades the following week’s contracts has to stay near the money; a wide spread on a later expiry stops with a clear error instead of trading at an invented price.

A positional spec

This bull put spread enters on a Supertrend flip up on 15-minute candles, needs at least three days to expiry, and holds across sessions until it reaches 60% of its credit, the day before expiry, or a Supertrend flip down, which closes it because there are no bearish legs to open:

{
  "name": "positional-bull-put-spread",
  "candle_minutes": 15,
  "entry": {
    "signal": { "kind": "supertrend_flip", "period": 10, "multiplier": "3" },
    "min_days_to_expiry": 3
  },
  "legs": [
    { "kind": "option", "option_type": "put", "side": "buy", "strike": { "kind": "atm_offset", "steps": 6 } },
    { "kind": "option", "option_type": "put", "side": "sell", "strike": { "kind": "atm_offset", "steps": 2 } }
  ],
  "exits": {
    "group_by": "expiry",
    "rules": [
      { "kind": "take_profit", "amount": { "kind": "share_of_credit", "fraction": "0.6" } },
      { "kind": "before_expiry", "days": 1 }
    ]
  }
}

There is no square_off, so the position is held overnight. The 15-minute candles are still built from one-minute data, so a fill after a 15-minute decision happens at the open of the next 15-minute candle, at the price recorded in that minute.

The MCX evening session

MCX gold and gold mini futures trade into the late evening, so an “intraday” futures strategy has a much longer day than an NSE one. Veloce’s futures trader allows entries for candles closing from 09:00 to 23:00 and squares off at 23:15 by default. The same rules about candles apply: build from one-minute data, decide on closes, fill at the next open. The MCX gold futures page covers the contracts.

Choosing a candle size for the test

If the strategy… Test on… Because
Has stops meant to be watched continuously 1-minute candles Closest to a stop order on the book
Enters at a specific minute 1-minute candles The entry and fill happen at that minute
Uses indicators on a 5 or 15 minute chart That candle size, built from 1-minute data The indicator must match the chart you trade from
Is positional with daily decisions 15 to 60 minute candles Fewer, slower decisions; gaps dominate anyway
Is uncertain Several sizes in one sweep The spread between them shows how much the candle matters

If the result changes sign between one-minute and five-minute candles, the strategy’s edge is inside the candle, and it will depend on exactly how you execute it.

Sizing and costs across both styles

Charges apply identically to both: NSE option charges on every fill, including STT at 0.15% of premium on sells for trades from April 1, 2026 (Finance Bill 2026 memorandum). Intraday strategies usually pay them more often. Lot size scales both: NIFTY went from 75 to 65 for weekly contracts from January 6, 2026 (NSE/FAOP/70616), and Veloce gives each contract the lot size for its own expiry, so a run that spans the change uses both.

Try it in Veloce

Veloce replays one-minute data for NIFTY 50 weekly and BANK NIFTY monthly options, SENSEX weekly options, and MCX gold and gold mini futures, and builds any candle size from it. Intraday strategies square off at the time you set; positional ones hold across sessions, exit before expiry, or settle at intrinsic value. The engine page explains the replay and fills, and the risk and exits page lists the exit rules.

Create a free Veloce account and run the same strategy on one-minute and five-minute candles to see how much the candle matters. For the full method, see how to backtest an options strategy, and for the timing rules that prevent lookahead bias, see the lookahead bias guide.

Frequently asked questions

What is the difference between intraday and positional trading in a backtest?
An intraday strategy closes everything before the session ends, so each day starts flat. A positional strategy holds across sessions, so the backtest has to handle overnight gaps, weekends, expiry and settlement. They are different strategies even when the entry rule is the same.
Why does a backtest need one-minute data?
Stops, fills and entry times are only as precise as the candles replayed. A five-minute candle can hide a spike that would have triggered a stop, and its next open can be minutes after the decision. With one-minute data you can still build any larger candle; you cannot go the other way.
Does Veloce check stops on every tick?
No. Veloce checks exits on each candle close of the strategy's candle size and fills at the next candle's open. On one-minute candles that is close to a stop watched every minute; real stop orders react to individual trades, so live fills can differ.
How does a positional backtest handle overnight gaps?
There are no candles overnight, so the first check after a gap is the close of the first candle of the next session, and the fill is at the open after that. A loss can be far larger than the stop level, which is what a real stop would also face.
Can I backtest the MCX evening session?
Yes. Veloce's MCX gold and gold mini futures backtests cover the evening session; the built-in futures trader allows entries until 23:00 and squares off at 23:15.

Sources

  1. NSE circular NSE/FAOP/68747: revision in expiry day of index and stock derivatives (June 25, 2025)nsearchives.nseindia.com, accessed
  2. NSE circular NSE/FAOP/70616: revision in market lot of derivative contracts on indices (October 3, 2025)nsearchives.nseindia.com, accessed
  3. Union Budget 2026-27, Memorandum explaining the provisions of the Finance Bill (STT on options)indiabudget.gov.in, accessed
esc
↑↓Move↵Open