Early access is open. Free sample datasets are live now. Download a sample

Data product · Early access

Crypto Tick Data: Trade-Level Download

Every executed trade on an exchange, with microsecond timestamps and the side that crossed the spread, so you can rebuild bars your own way and see the order flow behind each move.

Coverage

What you get

Item Detail
Product code CR-TRADES
Coverage Spot and perpetual pairs on the exchanges we carry
Granularity One row per trade, timestamps to the microsecond or finer
History Starter: last 30 days by REST. Pro: full history plus the live stream
Delivery REST, WebSocket, CSV or Parquet files
Update speed Live on Pro; daily files

Below about $79 a month, the market mostly sells live trade streams, not trade history. Above that, some sellers won't sell you a single pair. We sell one pair for one month as a file, at $3, and you keep it.

Schema

Schema

Field names match our published sample files.

Field Type Required Description
symbol string yes Pair, e.g. BTC-USDT
exchange string yes Venue code, e.g. binance
trade_id string yes Exchange trade id
timestamp string yes Exchange trade time, ISO 8601 UTC, up to nanoseconds
received_timestamp string optional When we received it
price number yes Trade price
size number yes Base-asset amount
notional number yes price times size, in quote currency
aggressor_side string yes buy or sell
sequence integer optional Exchange sequence number
is_liquidation boolean optional Trade was a forced liquidation

Example record, based on the first row of the BTC sample:

{
  "symbol": "BTC-USDT",
  "exchange": "binance",
  "trade_id": "6809266910",
  "timestamp": "2026-11-01T00:00:00.447286Z",
  "received_timestamp": "2026-11-01T00:00:00.451102Z",
  "price": 77582.02,
  "size": 0.00105,
  "notional": 81.46,
  "aggressor_side": "sell",
  "sequence": null,
  "is_liquidation": false
}

Sample data

Sample data

Bitcoin Tick Trades. The first 500 BTC/USDT trades of 1 November 2026. First 10 rows:

symbol trade_id timestamp price size notional aggressor_side
BTC-USDT 6809266910 2026-11-01T00:00:00.447286Z 77582.02 0.00105 81.46 sell
BTC-USDT 6809266911 2026-11-01T00:00:00.452197Z 77582.02 0.00217 168.35 sell
BTC-USDT 6809266912 2026-11-01T00:00:00.740755Z 77582.03 0.00025 19.4 buy
BTC-USDT 6809266913 2026-11-01T00:00:00.879536Z 77582.02 0.0797 6183.29 sell
BTC-USDT 6809266914 2026-11-01T00:00:00.879585Z 77582.02 0.01872 1452.34 sell
BTC-USDT 6809266915 2026-11-01T00:00:00.879799Z 77582.02 0.0234 1815.42 sell
BTC-USDT 6809266916 2026-11-01T00:00:00.905006Z 77582.02 0.00009 6.98 sell
BTC-USDT 6809266917 2026-11-01T00:00:01.071401Z 77582.03 0.00036 27.93 buy
BTC-USDT 6809266918 2026-11-01T00:00:01.208504Z 77582.02 0.0001 7.76 sell
BTC-USDT 6809266919 2026-11-01T00:00:01.210645Z 77582.03 0.0026 201.71 buy

Source: Binance

Ethereum Tick Trades. The first 500 ETH/USDT trades of the same day, same columns. Download ZIP (500 rows, 11.5 KB)

Source: Binance

Notice the pattern in the BTC rows: sells print at 77582.02 and buys at 77582.03. That one-cent gap is the spread, and aggressor_side tells you which side of it each trade hit.

Use cases

Use cases

Custom bars. Build volume bars, dollar bars or 10-second bars from price and size instead of being stuck with fixed time intervals. Dollar bars, summed on notional, give a steadier sample across quiet and busy hours.

Cumulative volume delta. Add size when aggressor_side is buy and subtract it when sell. A rising delta while price stalls points to buyers absorbing supply, or the reverse.

Large trade detection. Filter rows where notional is above your threshold, say $250,000, and plot them on the price chart. Clusters of big aggressive sells often show up before a breakdown.

Execution research. Compare your own fills with the trades printed around them by timestamp. That shows whether you paid the spread or captured it.

Code

Code examples

Example: one hour of ETH trades.

curl "https://api.marketdata.co/v1/crypto/trades?symbol=ETH-USDT&exchange=binance&start=2026-11-01T00:00:00Z&end=2026-11-01T01:00:00Z" \
  -H "Authorization: Bearer $MARKETDATA_API_KEY"

Example: Python, cumulative volume delta from the free sample.

import pandas as pd

t = pd.read_csv("btc-usdt-trades.csv", parse_dates=["timestamp"])
t["signed"] = t["size"].where(t["aggressor_side"] == "buy", -t["size"])
t["cvd"] = t["signed"].cumsum()
print(t[["timestamp", "price", "aggressor_side", "cvd"]].tail())

Pricing

Pricing

Plan Crypto trades
Starter, $29/mo REST trades for the last 30 days
Pro, $79/mo Live trade WebSocket and full history
Business, $249/mo Commercial use

Dataset: $3 per pair per month; a whole exchange for one day at $9. The first day of each month is planned as a free file. Among the providers we studied, free tiers exist, and paid plans run from $29.99 to $1,500 a month, median $79.

Side by side

How we compare

12 of the 50 providers we studied sell crypto trades. Cheaper plans tend to be live streams only, and the deep historical archives start at hundreds of dollars a month and won't sell single instruments. Per-pair, per-month files with a listed price fill that gap, and they use the same symbol and timestamp format as our bars.

FAQ

Frequently asked questions

What does aggressor_side mean?

It is the side that crossed the spread. buy means a buyer took an ask, sell means a seller hit a bid. Exchanges call it the taker side.

How precise are the timestamps?

The sample has microseconds, and the format supports up to nanoseconds where the exchange provides them. All times are UTC.

Are trades ever corrected or removed?

If an exchange corrects or cancels a trade, we follow its published record. trade_id lets you match rows across downloads and spot any change.

Can I get one pair for one month?

Yes. That is the unit we sell: $3 per pair per month, as CSV or Parquet, yours to keep.

Is it real-time?

The live trade stream comes with Pro. Starter gives REST access to the last 30 days.

Can I use the Binance samples commercially?

Check Binance's terms before commercial display. Paid data carries commercial internal use from Business.

Keep reading