Schema
The five fields
For each bar (one day, one hour, one minute):
- Open: the first trade price in the period.
- High: the highest trade price.
- Low: the lowest trade price.
- Close: the last trade price.
- Volume: the total quantity traded, in shares, contracts or coins.
Our bars add a few fields that answer common follow-up questions: vwap (volume-weighted average price within the bar), trade_count, and for crypto quote_volume (volume in the quote currency, here USDT) and taker_buy_volume (the part of volume where the buyer was the aggressor).
Three real daily bars
From our free BTC and ETH daily file, the first three days of July 2026 for BTC-USDT:
| symbol | interval | timestamp | open | high | low | close | volume | vwap | trade_count |
|---|---|---|---|---|---|---|---|---|---|
BTC-USDT |
1d | 2026-07-01T00:00:00Z | 58624.71 | 61334 | 57800.19 | 60024 | 25093.44662 | 59419.95 | 4166849 |
BTC-USDT |
1d | 2026-07-02T00:00:00Z | 60024 | 62200 | 59588 | 61560 | 21382.14163 | 61179.87 | 3883096 |
BTC-USDT |
1d | 2026-07-03T00:00:00Z | 61560 | 62979.86 | 61248.86 | 62583.26 | 14048.98627 | 61951.8 | 2571531 |
Source: Binance. Columns quote_volume and taker_buy_volume omitted here.
What you can read straight off these rows:
- Return: 1 July closed at 60,024 and 2 July at 61,560, a gain of 2.56%.
- Range: 1 July's high minus low was 3,533.81, about 6.0% of the open. 3 July's range was only 1,731.
- Close vs VWAP: on all three days the close was above the bar's VWAP. Late buyers paid more than the day's average, a sign of strength into the close.
- Activity: volume and trade count fell each day, from 4.17 million trades to 2.57 million, even as price rose.
Notice the open of each day equals the previous close. That is normal for a market that trades around the clock. US stocks are different: the daily open is the first trade after 9:30 ET and can gap far from the prior close.
Details
Load and use OHLCV in pandas
"""Example: returns, true range and gaps from daily OHLCV bars."""
import pandas as pd
df = pd.read_csv("btc-eth-daily-bars.csv", parse_dates=["timestamp"])
btc = df[df["symbol"] == "BTC-USDT"].set_index("timestamp").sort_index()
btc["ret_pct"] = btc["close"].pct_change() * 100
prev_close = btc["close"].shift(1)
btc["true_range"] = pd.concat([btc["high"] - btc["low"],
(btc["high"] - prev_close).abs(),
(btc["low"] - prev_close).abs()], axis=1).max(axis=1)
btc["atr14"] = btc["true_range"].rolling(14).mean()
btc["gap_pct"] = (btc["open"] / prev_close - 1) * 100
btc["buy_share"] = btc["taker_buy_volume"] / btc["volume"]
print(btc[["close", "ret_pct", "atr14", "gap_pct", "buy_share"]].tail(5).round(3))
atr14 is the average true range over 14 days, a common way to size stops. gap_pct will be near zero for this crypto file; on stock data it measures the overnight gap. buy_share above 0.5 means aggressive buyers did more than half the volume that day.
Watch out
Common mistakes
Assuming the high came before the low (or after). A bar doesn't record the order. A backtest that needs to know whether a stop or target hit first can't get that from OHLCV alone.
Mixing up bar start and bar end timestamps. Ours mark the start of the bar. Some vendors stamp the end. Shift by one bar and your signals use data from the future.
Comparing volume across asset types. Stock volume is shares, futures volume is contracts, crypto volume is coins. Use notional (quote_volume for crypto) when you compare across markets.
Ignoring adjustments on stocks. A stock's raw OHLC history jumps at every split. Use split-adjusted bars for anything that spans a split.
Details
Get the data
Free OHLCV in our standard schema. Daily and 1-minute crypto bars, with
vwap,trade_countandtaker_buy_volumeon every row.
- Download BTC and ETH daily bars (246 rows, July to October 2026)
- Download altcoin daily bars (615 rows: SOL, XRP, BNB, DOGE, ADA)
- Source: Binance
- Stocks: historical daily stock data, available in early access. A stock sample is available on request.
FAQ
Frequently asked questions
What is OHLCV data?
OHLCV data is price data summarised into bars, each with the open, high, low and close price and the volume traded in that period. It is the standard input for charts, indicators and most backtests.
What is the difference between OHLC and OHLCV?
OHLCV adds volume. Some sources, especially spot FX, provide OHLC without true volume because there is no central exchange, and may offer a tick count instead.
What time period does a daily bar cover?
For US stocks, the regular session from 9:30 to 16:00 ET, sometimes with extended hours depending on the vendor. Crypto bars usually run 00:00 to 24:00 UTC. Always check the vendor's definition.
Is the close the same as the settlement price?
Not always. For stocks the official close comes from the closing auction. In futures, the settlement price is set by the exchange by its own method and can differ from the last trade.
Keep reading