Historical Data Backtesting: Tick, Minute, and Daily Data Compared
Blog by Aliceblue News Desk
30 October 2026
13 minutes
You tested a trading strategy on old charts, and it looked profitable. Then you took it live, and the results changed. One reason could be the historical data you used. A daily chart, a 1-minute chart, and tick-by-tick data do not show the market in the same way.
Historical data backtesting means checking your trading rules against past prices before risking money. Daily data works for swing and positional trades, minute data suits most intraday strategies, and tick data is useful only when your stop loss or target depends on very small price moves.
Tick data records every trade or price update. Minute data gives one open, high, low, and close candle every minute. Daily data gives one candle per trading day. Use daily data for swing trades, minute data for intraday trades, and tick data when your entry, stop loss, or target sits within a very small price range.
What Is Historical Data for Backtesting?
What Is Historical Data for Backtesting?
Historical data is the record of how a stock, index, futures contract, or option moved in the past. It usually includes price, volume, and time.
Backtesting means applying your trading rules to this old price history. For example, if your rule is “buy Nifty futures when price crosses above the first 15-minute high,” a backtest checks how that rule would have worked in earlier sessions.
The difference between tick, minute, and daily data is the level of detail.
- Tick data records every trade or price update.
- Minute data records one candle for every minute.
- Daily data records one candle for the full trading day.
A backtest can only be as good as the data behind it. If the data misses price moves or shows incorrect prices, the results may not reflect what would have happened in real trading.
How Does Historical Data Backtesting Work?
- Write your rules clearly. For example: “Buy Bank Nifty futures when price breaks the first 15-minute high. Exit at a 40-point profit or a 25-point loss.”
- Choose data that matches your strategy. If you trade on 5-minute candles, minute data is enough. If your stop loss is only a few points away, minute data may not show which price level was touched first.
- Clean the data. Check for missing candles, wrong prices, and unadjusted stock prices. A 1:5 stock split can look like an 80 percent fall if the data is not adjusted.
- Run the rules candle by candle. The backtesting software checks each candle and records every entry, exit, profit, and loss.
- Add real costs. Include brokerage, Securities Transaction Tax (STT), exchange charges, GST, stamp duty, and slippage. Slippage is the difference between the price you expected and the price you actually received.
- Check more than profit. Look at the number of trades, win rate, losing streaks, and maximum drawdown. Maximum drawdown is the biggest fall in your capital from an earlier high.
- Test on unseen data. Build the strategy on one period and check it on another. This reduces the chance that you have only fitted rules to one particular market phase.
What Is Tick Data in Backtesting?
Tick data records individual market events. Depending on the source, it may include every trade, every quote update, or both. A tick usually carries a timestamp, price, and volume. Some datasets also include the best bid and ask prices.
The bid is the price buyers are willing to pay. The ask is the price sellers want. The gap between them is called the spread.
Tick data shows the exact order in which prices moved. If a stock touched ₹1,000, rose to ₹1,001 and fell back to ₹1,000 within seconds, tick data can show that sequence.
This matters when your strategy depends on small price differences. Scalpers, options traders, and algo traders often need this detail because a few points can decide whether a trade hits the target or the stop loss.
The trade-off is size and cost. Tick files can become very large, especially for highly liquid stocks and indices. They also need more storage, processing time, and technical skill.
What Is Minute Data in Backtesting?
Minute data groups all price activity in one minute into a single candle. Each candle shows:
- Opening price
- Highest price
- Lowest price
- Closing price
- Total traded volume
For example, a 1-minute candle for Reliance Industries may show an open of ₹2,900, a high of ₹2,904, a low of ₹2,898, and a close of ₹2,902.
Minute data is practical for most intraday traders. You can build 1-minute, 3-minute, 5-minute, and 15-minute strategies from it.
Its limitation is that it does not show the order of prices inside the candle. If both your target and stop loss fall within one candle, the candle alone cannot tell you which was hit first.
What Is Daily Data in Backtesting?
Daily data gives one candle per trading day. It shows the day’s opening price, highest price, lowest price, closing price, and total traded volume.
This is the easiest data to understand and use. Many charting platforms and broker platforms provide daily price history, though the date range and export options can vary.
Daily data works well for:
- Swing trading
- Positional trading
- Long-term investing research
- Strategies based on moving averages, weekly breakouts or monthly patterns
It does not work for intraday strategies. A daily candle cannot tell you whether a stop loss was hit at 10:15 am or whether price recovered by 2:30 pm.
NSE’s regular equity market session runs from 9:15 am to 3:30 pm, while its equity derivatives session runs from 9:15 am to 3:40 pm.nseindia
Which Historical Data Should You Use for Backtesting?
The right data depends on how closely your strategy watches price. A useful rule is: your data should be at least as detailed as the timeframe on which you make decisions.
Your strategy | Best data | Why |
Hold for days, weeks or months | Daily | One candle per day is enough |
Trade on 5-minute or 15-minute charts | Minute | You need intraday highs and lows |
Trade on 1-minute charts | Minute, sometimes tick | Minute data works unless stops are very tight |
Scalp with 2 to 10 point targets | Tick | You need the exact order of price moves |
Backtest options strategies | Minute or tick, plus option chain data | Index candles alone are not enough |
Study long-term stock performance | Daily | Easy to access and covers more years |
Another simple check is stop loss distance. If your stop loss is 0.5 percent away from entry, minute data is usually enough. If it is only a few points away, tick data becomes important.
Practical Example: Why Minute Data Can Mislead
Suppose you test this rule on Nifty futures:
- Buy at 24,500
- Target: 24,530
- Stop loss: 24,480
- Lot size: 65 units
The Nifty 50 futures lot size is 65 units for contracts from the January 2026 cycle. It was revised from 75 units.ianslive+1
One futures lot represents a fixed number of units. With a 65-unit Nifty futures lot, every 1-point move changes your profit or loss by ₹65.
If the target is hit:
- Profit = 30 points × 65 = ₹1,950
If the stop loss is hit:
- Loss = 20 points × 65 = ₹1,300
Now imagine one 1-minute candle shows:
- Open: 24,500
- High: 24,531
- Low: 24,479
- Close: 24,510
Both your target and stop loss are inside this candle. But the candle does not tell you which price came first.
A minute data backtest has to make an assumption. If it assumes the target was hit first, it records a ₹1,950 profit. If it assumes the stop loss was hit first, it records a ₹1,300 loss.
That is a difference of ₹3,250 on one trade.
Now add costs. Suppose brokerage, taxes, exchange charges, and slippage come to ₹300 for a round trip.
- Target profit after costs: ₹1,650
- Stop loss loss after costs: ₹1,600
If your backtest ignores these costs, it may show a profit that does not exist in live trading.
This example assumes the order fills exactly at the trigger price. In live trading, slippage can change the entry, stop loss, or target price.
Tick data removes the guesswork because it shows the sequence of prices. But for a strategy with a 100-point target and a 50-point stop, minute data will usually give a similar result at much lower cost.
Why Options Backtesting Needs Option Data
Many traders try to backtest Nifty or Bank Nifty option strategies using only index candles. That does not work properly.
Nifty index candles show the index price. They do not show:
- Option premium
- Strike-wise call and put prices
- Implied volatility (IV)
- Open interest
- Bid-ask spread
- Futures premium or discount
- Expiry day price behaviour
Implied volatility is the market’s expectation of future price movement. It affects option premiums, so two options with the same strike can behave very differently when IV changes.
Open interest is the number of option contracts still active.
For example, if your strategy buys a Nifty 24,500 CE, you need the historical price of that exact option contract. You also need to know how its premium moved during the trade. Historical option chain data includes strike-wise premiums, open interest, volume, and IV, which is why it is needed for option backtesting.
Tick vs Minute vs Daily Data for Backtesting
Parameter | Tick Data | Minute Data | Daily Data |
What it records | Every trade or price update | OHLC and volume per minute | OHLC and volume per day |
Detail level | Highest | Medium | Lowest |
Shows order of price moves | Yes | No | No |
File size | Very large | Moderate | Small |
Best for | Scalping, algo trading, tight stops | Intraday and short-term trading | Swing, positional and investing |
Typical history available | Often shorter | Several years | Many years |
Cost | High | Moderate | Low or free |
Beginner friendly | No | Yes, after basics | Yes |
How Much Historical Data Do You Need for Backtesting?
There is no fixed number of years. You need enough trades and enough different market conditions.
For daily strategies, five to ten years is a good starting point. This lets you see how the strategy behaved during rising markets, sharp falls, and long sideways phases. Recently listed companies will have less history, so the test period will naturally be shorter.
For intraday strategies, focus on the number of trades. A strategy with only 30 trades over three months does not give a reliable result. Test it across trending days, range-bound days, and high volatility days.
For options strategies, include different expiry cycles. Weekly and monthly options behave differently because time decay and volatility change as expiry approaches.
Data Quality Checklist for Indian Markets
Before trusting a backtest, check these points:
- Adjusted prices: Stock data should account for splits, bonuses, dividends, rights issues and demergers. Otherwise, old charts can show artificial price gaps.
- Missing candles: Check whether the data has gaps during regular market hours.
- Market holidays: The data should correctly exclude non-trading days.
- Index vs futures: Nifty index data is not the same as Nifty futures data. Futures can trade at a premium or discount to the index.
- Expired contracts: Futures and options contracts expire. Make sure you are testing the contract that was actually available on that date.
- Point-in-time data: Use the information that was available on that date. Do not use today’s company list, contract details, or adjusted prices to judge a past trade.
- Low liquidity stocks: A strategy may look good in backtesting but fail live if there are not enough buyers and sellers.
- Costs: Include brokerage, STT, exchange transaction charges, SEBI charges, GST, and stamp duty.
- Slippage: Add a buffer for orders that fill at a worse price than expected.
- Source reliability: Cross-check a few dates against another source. If the high, low, or close does not match, investigate before using the data.
Corporate action adjustment is especially important for equity backtesting. Splits, bonus issues and dividends can leave visible jumps in raw price history, while adjusted series rescale older prices to remove those artificial gaps.
Advantages and Risks
Advantages
- Better strategy testing. The right data level gives results closer to what you may see in live trading.
- Lower cost to start. Daily data is often free, so beginners can learn backtesting without buying expensive feeds.
- Faster testing. Minute and daily data are easier to process than tick data.
- Better risk control. A proper backtest shows drawdowns, losing streaks, and trade frequency before you risk money.
- Useful for algo trading. If you plan to automate a strategy, historical testing helps you check the rules before connecting them to a broker API.
Risks and Limitations
- Past performance does not repeat exactly. Market behaviour, liquidity, and volatility change.
- Bad data creates false confidence. Missing candles, wrong prices and unadjusted corporate actions can distort results.
- Costs can turn a profitable strategy negative. This is especially true for strategies with many small trades.
- Slippage is often ignored. In fast-moving markets, your order may not fill at the exact price shown on the chart.
- Overfitting is common. If you keep changing rules until old data looks perfect, the strategy may fail live.
- Survivorship bias can mislead you. If you test only companies listed today, you leave out companies that were removed from the exchange after failing.
- Index data is not tradable. Nifty index candles do not include futures premium or option prices.
Common Mistakes in Historical Data Backtesting
- Using daily data for intraday rules. Daily candles cannot show intraday stop-loss hits. Use minute data instead.
- Ignoring brokerage and taxes. Always include all charges before judging a strategy.
- Assuming perfect fills. Add a slippage buffer, especially for options and low-volume stocks.
- Testing only one market phase. A strategy that works in a rising market may fail when the market falls.
- Using too few trades. Ten profitable trades do not prove a strategy works.
- Changing too many settings. Keep rules simple. More settings usually mean more overfitting.
- Using unadjusted stock prices. Splits, bonuses, and dividends can create fake price gaps.
- Buying tick data too early. First test with minute data. Move to tick data only if the results change materially.
- Testing options with index data only. You need historical option chain data to test option premiums correctly.
Tips for Beginners
- Start with daily data and one simple rule.
- Move to minute data only when you trade intraday.
- Keep stop losses and targets wider than the normal range of one candle.
- Include every cost in the backtest.
- Test the strategy in rising, falling, and sideways markets.
- Keep some data aside for final testing.
- Write down every rule change you make.
- Paper trade the strategy or use a very small position size before trading with full capital.
- Compare results on minute and tick data before paying for tick history.
Who Should Use Tick, Minute or Daily Data?
Daily data suits investors, positional traders, and swing traders. It is also the best place for a beginner to learn backtesting.
Minute data suits intraday traders, breakout traders, and traders who use 1 minute, 5 minute or 15-minute charts.
Tick data suits scalpers, high-frequency traders, and algo traders whose profit per trade is small. It is less useful for long-term investors and traders with wide stop losses.
If you are new, start with daily or minute data. Tick data is useful, but it adds cost, storage needs, and technical work. Use it only when your stop loss or target is small enough to make that detail matter.
Conclusion
For historical data backtesting, tick, minute, and daily data serve different purposes. Daily data works for slower strategies, minute data covers most intraday rules, and tick data is needed only for very tight price levels. Start with the simplest data that answers your question, clean it properly, add real costs, and test across different market conditions. Then test the strategy with a small position before increasing exposure.
Frequently Asked Questions
Blog by Aliceblue News Desk
