Coin Metrics is pleased to announce the version 2.6 release of our CM Market Data Feed. This release contains several new features, upgrades, and bug fixes.

New Features

  • Uniswap and Sushiswap swaps, liquidity pool metadata, candles, and volume metrics. Coin Metrics has been actively working on collecting data from major defi protocols. For our first release, we are adding support for all major liquidity pools on Uniswap v2, Uniswap v3, and Sushiswap v1.

    Swaps data is served through our existing /timeseries/market-trades endpoint because swaps are conceptually identical to a trade. Users can see all the standard trade fields for a swap such as time, price, and volume but can also see defi-specific fields such as the block hash, transaction id, the addresses involved in the swap, and more by passing the include_defi_fields=true parameter.

    Metadata about all major liquidity pools such as the base and quote asset, contract addresses, fees are also available through our /catalog/markets endpoint.

    Candles for all major liquidity pools are also available through our /timeseries/market-candles endpoint.Reported spot volume metrics (volume_reported_spot_usd_1d and volume_reported_spot_usd_1h) are available at the exchange and exchange-asset level through our /timeseries/exchange-metrics and /timeseries/exchange-asset-metrics endpoints. Market Data Feed clients can access the data by contacting info@coinmetrics.io for assistance. DeFi protocol data has been in high demand, so we are excited to make this available in the form of a free, early-access trial. As we eventually expand our coverage to include major AMMs and lending protocols, the complete DeFi package may become a separate add-on at a later date. As we continue in our development efforts, we encourage your feedback on the data and other protocols that you’d like us to expand to.
  • Serving data in flat files. In addition to our API, we now serve some data types through daily flat files. The currently supported data types are tick-by-tick quotes for all markets on major exchanges, trades for all markets, and candles for all markets. These flat files are useful for users who would like to backfill large amounts of history or consume or data using daily batch jobs. Market Data Feed clients can access the data by contacting info@coinmetrics.io.
  • Liquidation metrics. We began collection of every single liquidation order or trade on major futures exchanges in 2020. Now we have begun publishing liquidation metrics in which individual liquidations are summed across some logical groupings to facilitate easier analysis. The following liquidation metrics are available through our /timeseries/exchange-asset and /timeseries/market-metrics endpoints:
    liquidations_reported_future_buy_units_5m
    liquidations_reported_future_buy_usd_5m
    liquidations_reported_future_sell_units_5m
    liquidations_reported_future_sell_usd_5m
    liquidations_reported_future_buy_units_1h
    liquidations_reported_future_buy_usd_1h
    liquidations_reported_future_sell_units_1h
    liquidations_reported_future_sell_usd_1h
    liquidations_reported_future_buy_units_1d
    liquidations_reported_future_buy_usd_1d
    liquidations_reported_future_sell_units_1d
    liquidations_reported_future_sell_usd_1d
  • Reference rates pair candles. We now serve candles quoted in U.S. dollars generated from our real-time reference rates for all assets in our reference rates coverage universe. The candles are available in intervals 1m, 5m, 10m, 15m, 30m, 1h, 4h, 1d. They are served through our new /timeseries/pair-candles endpoint.
  • Streaming market candles. Our candles are derived directly from trades data that we collect from exchanges. Previously, we waited 20 minutes before calculating candles to ensure that all trades in an underlying interval are collected. Now we are serving streaming candles calculated in real-time with no delay from our new /timeseries-stream/market-candles endpoint. The intervals 1m, 5m, 10m, 15m, 30m, 1h, 4h, 1d for all spot and futures markets in our coverage universe.
  • Reference Rates release. We previously published our release notes for our Hourly Reference Rates v2.11 and Real-Time Reference Rates v0.11. This release includes a regularly scheduled reconstitution of the constituent markets for all assets, a modification of the methodology regarding the eligible markets that are considered to serve as constituent markets, the addition of 94 additional cryptocurrencies to our coverage universe and some terminations to our coverage universe.
  • Expanded order book coverage. We now cover snapshots of the top 100 bids and asks as well as full order book snapshots of OKEx spot markets, FTX.US spot markets, and CME futures markets. We also added more frequent order book snapshots for several highly traded perpetual futures across many futures exchanges. We support spot order books from 19 exchanges and futures order books from 12 exchanges. To find out the latest order book coverage universe, please contact us.
  • Expanded quotes coverage. Our quotes coverage now matches our order book coverage. We now serve hourly quotes snapshots for all markets on all major spot, future, and option exchanges through our /timeseries/market-quotes endpoint. We support spot order books from 19 exchanges and futures order books from 12 exchanges. To find out the latest quotes coverage universe, please contact us. We continue to serve 10 second snapshots of a subset of major markets, and streaming quote and order book data is still available through our websocket API.
  • Support for HitBTC futures markets. Trades, order book, quotes, candles, futures contract specifications, and volume metrics for HitBTC futures markets are available.
  • Support for FTX.US spot markets. Trades, order book, quotes, candles, and volume metrics for FTX.US spot markets are available.

Upgrades

  • More order book depth options. We added a depth_limit parameter to our /timeseries/market-order-book endpoint where users can specify how many order book levels to return. Supported values are 1-30000 or full order book.
  • Support for market_per_limit parameter for all market data endpoints. The market_per_limit parameter is useful for getting a set amount of data for a wide selection of markets, such as the most recent observation. Now this parameter is available for all market (/timeseries/market-*) endpoints.
  • Added include and exclude parameters for catalog endpoints. Some of our catalog responses are large, so we added an include and exclude parameter to some of them so that users can specify exactly which fields to include or exclude. It is available for /catalog/assets, /catalog/markets, /catalog-all/assets, and /catalog-all/markets endpoints.
  • Improved candle logic that converts volume to U.S. dollars. Our candles contain a candle_usd_volume field which represents the volume in U.S. dollars. We have improved the conversion logic we use so that it selects higher quality reference rates to perform the conversion so that this field is even more accurate.
  • Improved our monitoring and ability to fill any gaps in our trades history.
  • Improved the latency of serving the order book updates served through our /timeseries-stream/market-orderbooks endpoint.
  • Improved the latency of our collection of some exchanges with extremely high trades activity like Coinbase and Binance.
  • Migrated our collection of Gate.io, Bibox, HitBTC, LBank, and Grayscale’s data to use the latest API version.
  • Added better handling of spot order book observations from itBit and Bybit where one side of the order book does not have orders.
  • Optimized the queries for several API calls.
  • Improved resiliency of reference rates served through our /timeseries/stream/asset-metrics endpoint.
  • Allowed the timestamps for our Binance open interest to conform to how we handle timestamps for other exchanges.

Bug Fixes

  • Fixed a bug that prevented order books from futures markets to be served through our /timeseries-stream/market-orderbooks endpoint.
  • Filled some gaps in our open interest metrics.
  • Fixed a small number of historical trades for CME Ethereum futures with incorrect trade price. Candles and volume metrics were also recalculated.
  • Candles for all spot markets for FTX.US spot markets were backfilled from 2020-04-05 through 2022-01-11.
  • Fixed an incorrect observation for our spot reported and trusted volume metrics on 2022-03-04.
  • Resolved some minor ticker collisions from Bibox and LBank.
  • Fixed some trades from FTX that were incorrectly assigned to FTX.US and recalculated downstream candles and volume metrics.
  • Fixed some trades from FTX.US that were incorrectly assigned to FTX and recalculated downstream candles and volume metrics.
  • Removed some anomalous trades from a small number of LBank markets. Recalculated downstream candles and volume metrics.
  • Fixed in instance where the API would return an error if a user asked for markets using the wildcard parameter like *-spot.
  • Fixed an instance where a small number of OKEx trades from certain markets were incorrectly assessed to be duplicates.
  • Removed an anomalous trade from LocalBitcoin on 2017-10-10 and recalculated downstream candles and volume metrics.

About CM Market Data Feed

The CM Market Data Feed provides access to historical and real-time data from the world’s leading spot and derivatives crypto exchanges. Our coverage universe currently consists of 3,117 assets, 38 exchanges, 17,250 spot markets, 9,139 futures markets, 45,309 options markets, 3,573 pairs, and one institution. We offer a wide range of harmonized datasets such as trades, candles, order book snapshots, futures-specific data types, options-specific data types, market data metrics, and more. More detailed and up-to-date information can be found at the Coin Metrics Data Encyclopedia.

Please contact Coin Metrics through our website or at info@coinmetrics.io for more information on the CM Market Feed.