Late-Arriving Market Data
Late-arriving market data is trade, quote, or order book information that is received after the time interval it belongs to has already closed.
Late-arriving market data can occur when trades, quotes, order book updates, or exchange corrections are delayed in transit or delivered after the close of the relevant period. In OHLCV datasets, late-arriving trades may affect the final open, high, low, close, or volume values for a candle that was already published in provisional form.
This is one reason historical market data systems distinguish between provisional values and finalized values. A candle may be available shortly after its period boundary, but late-arriving market data can still cause that candle to be updated during a reconciliation or correction window.
Late-arriving data is especially important in cryptocurrency markets because exchanges, symbols, and data feeds can have different latency, outage, and correction behavior. Data providers need clear rules for how late data is incorporated into historical datasets and when a record should be treated as final.
Why Late-Arriving Market Data Matters
Late-arriving market data matters because it can change historical values that were already consumed by trading systems, dashboards, risk models, or backtests. If downstream systems assume a candle is final too early, they may store incomplete or incorrect OHLCV values.
For analytics teams, late-arriving data affects data quality, reproducibility, and historical consistency. For traders and developers, it influences whether a recently closed candle should be used for final reporting, strategy signals, or backtesting.
Clear handling of late-arriving data helps users understand when to query updated historical data, when to run backfills, and how to design systems that tolerate corrections.
How Does Late-Arriving Market Data Affect OHLCV Candles?
Late-arriving trades can modify an OHLCV candle after the candle interval has closed. A delayed trade may change the high, low, close, or volume of the period, depending on its timestamp and price.
For example, if a one-minute BTC/USD candle closes at 10:01:00 but a valid trade with a timestamp of 10:00:45 arrives at 10:01:05, the 10:00:00-10:01:00 candle may need to be recalculated.
Is Late-Arriving Market Data the Same as Backfill?
Late-arriving market data and backfill are related but not identical. Late-arriving data refers to delayed information that belongs to an earlier time period. Backfill is the process of adding or correcting historical records after the fact.
Late-arriving data may trigger a backfill, but backfills can also be caused by exchange outages, data normalization changes, symbol mapping updates, or corrections discovered later.
How Should Applications Handle Late-Arriving Market Data?
Applications should treat very recent OHLCV candles as potentially provisional unless the data provider indicates that the candle is final. Systems that require historical accuracy should support re-querying recent intervals, applying corrections, and recording when data was first published or later updated.
Backtesting and reporting workflows may also benefit from using dataset versions or availability timestamps so results can be reproduced with the same data vintage.
Real-World Example
A trading dashboard requests one-minute ETH/USD candles immediately after each minute closes. At 12:01:01, it receives a candle for 12:00:00-12:01:00. A few seconds later, the exchange sends a delayed trade with an execution timestamp of 12:00:58. The candle's close and volume are recalculated, and the historical OHLCV endpoint later returns the corrected candle.
In this case, the delayed trade is late-arriving market data, and the updated candle is the result of processing that data.
Late-Arriving Market Data and CoinAPI Market Data API
CoinAPI Market Data API helps developers access normalized real-time and historical cryptocurrency market data across many exchanges. When working with OHLCV candles, trades, or historical datasets, understanding late-arriving market data helps developers design systems that can handle provisional candles, corrections, reconciliation windows, and backfills.
Applications that depend on precise historical candles should consider rechecking recent intervals and using consistent data retrieval policies for analytics, reporting, and backtesting.