Tick Data and OHLCV Data
Tick data records every trade or quote; OHLCV bars summarise them by time. Learn how bars are built, other bar types, storage costs and which data a strategy needs.
Market data comes at different resolutions. Tick data records every individual trade, and sometimes every quote change, exactly as it happened. OHLCV bars summarise those ticks into fixed periods: the open, high, low and close prices plus volume for each minute, hour or day. Bars are compact and easy to work with, which is why most charts and backtests use them. Ticks preserve details that bars throw away, such as the order in which highs and lows happened. Choosing the right resolution affects both the cost of research and whether its results can be trusted.
What each contains#
| Tick data | OHLCV bars | |
|---|---|---|
| Content | Each trade (time, price, size), and for quote data each bid and ask change | Open, high, low, close and volume per period |
| Size | Very large | Much smaller |
| Detail | Full sequence of events | Summary per period |
| Best for | Execution research, high frequency, precise fills | Most indicators, swing and position strategies |
How a bar is built#
For each period, take all trades inside it:
- Open: price of the first trade.
- High: highest trade price.
- Low: lowest trade price.
- Close: price of the last trade.
- Volume: sum of trade sizes.
This is exactly what a candlestick shows. See How to Read Candlesticks.
Bar labelling: start or end?#
Some vendors label a bar with its start time (10:00 for 10:00 to 10:01); others use the end time. If you treat a start labelled bar as complete at its label time, your backtest sees its close a full bar early. Always check the convention. See Timestamps, Time Zones and Daylight Saving and Look-Ahead Bias.
Other bar types#
| Bar type | New bar forms when | Why use it |
|---|---|---|
| Time bars | A fixed time passes | Standard and simple |
| Tick bars | A set number of trades occur | More bars in active periods |
| Volume bars | A set volume trades | Normalises activity |
| Dollar bars | A set value trades | Comparable across price changes |
| Range and Renko bars | Price moves a set amount | Focus on movement, not time |
Some researchers argue that volume and dollar bars produce returns with better statistical properties than time bars, since they sample more during busy periods.
Storage costs#
Tick data can be hundreds or thousands of times larger than minute bars. A liquid futures contract can print hundreds of thousands of trades in a busy day, and quote data is larger still. Columnar formats and compression help a lot. See Data Storage, Compression and Caching.
Which data do you need?#
| Strategy | Minimum data |
|---|---|
| Long term investing, monthly rebalancing | Daily bars |
| Swing trading | Daily or hourly bars. See Swing Trading |
| Intraday trading with stops and targets | Minute bars, with tick checks on fill sensitive rules |
| Scalping and market making | Tick and quote data. See Scalping |
| Execution research | Trades and quotes with precise timestamps |
Quotes versus trades#
Trade prices tell you what happened; quotes tell you what you could have traded at. A backtest using last trade prices for entries ignores the spread. For short term strategies, use bid and ask data to model buying at the ask and selling at the bid. See Bid-Ask Spread and Spread Costs.
Frequently asked questions#
What is tick data?#
A record of every individual trade, and sometimes every quote change, with its time, price and size.
What does OHLCV stand for?#
Open, high, low, close and volume, the summary values of a price bar.
Do I need tick data for backtesting?#
Not for most daily or hourly strategies. Short term strategies with tight stops and targets, or that depend on the spread, benefit from tick and quote data.
Next, learn how live data reaches your system in Real-Time, Delayed and Historical Data.
3 quick questions on this lesson. Get them all right to finish it.
Turn on JavaScript to take the quiz.
Mentioned in
- NumPy and Pandas for TradersProgramming and Data
- WebSocket Market Data StreamsProgramming and Data
- SQL for Trading DataProgramming and Data
- Market Data Levels: Level 1, 2 and 3Programming and Data
- Quant Trading Learning PathStart Here
- Ticks and Tick SizeMarkets and Instruments