Availability Time
Availability time is the moment when market data becomes accessible to users or systems after it has been collected, processed, validated, and published.
Availability time is the point at which market data becomes available for consumption through an API, feed, file, or dataset. It describes when users can actually access a record, not necessarily when the underlying market event happened.
For cryptocurrency market data, availability time is important because trades, quotes, order book updates, and OHLCV candles may be collected and processed before they are exposed to downstream systems. The difference between event time and availability time can affect backtesting, alerting, analytics, and real-time trading workflows.
What is availability time?
Availability time is the timestamp or moment when a data provider makes a piece of market data accessible to users, applications, or downstream systems.
It is different from the time of the original market event. A trade may occur at one time, be received by a data platform shortly afterward, and become available through an API after normalization, validation, storage, or aggregation steps are complete.
In market data systems, availability time helps answer a practical question: when could a user have known this data?
Why does availability time matter in market data?
Availability time matters because financial systems often depend not only on what happened, but also on when the information became available.
For example, a backtest that uses data before it would have been available in real life can introduce look-ahead bias. A trading system that expects newly closed OHLCV candles immediately may need to account for small delays, late-arriving records, or post-close reconciliation.
Tracking availability time helps teams build more realistic analytics pipelines, understand latency, and separate historical truth from real-time data availability.
How is availability time different from event time?
Event time is when the market event occurred, such as the timestamp of a trade or quote update. Availability time is when that event became accessible from a provider or dataset.
For example, a trade might occur at 12:00:01, be received by a market data platform at 12:00:02, and become available in an API response at 12:00:04. In that case, the event time is 12:00:01, while the availability time is 12:00:04.
This distinction is especially important when working with real-time feeds, historical datasets, and reconstructed datasets where records may be corrected or backfilled later.
What affects availability time?
Availability time can be affected by exchange connectivity, network latency, data normalization, validation, storage, aggregation, reconciliation, and publication workflows.
In OHLCV data, availability time may also depend on candle construction. A candle for a completed period may not be considered available until trades for that interval have been processed and the candle has been published through the relevant API or dataset.
Late-arriving market data, correction windows, and backfill processes can also change when corrected or more complete versions of records become available.
Real-world example of availability time
Suppose a one-minute OHLCV candle covers the interval from 10:00:00 to 10:01:00. The candle period closes at 10:01:00, but the final candle is published through a market data API at 10:01:03 after the relevant trades have been processed.
In this case, the candle's period boundary is 10:01:00, but its availability time is 10:01:03. A backtest that assumes the candle was usable at exactly 10:01:00 may be more optimistic than a system that respects when the candle was actually available.
Availability Time and CoinAPI Market Data API
CoinAPI Market Data API provides access to normalized cryptocurrency market data across exchanges, including trades, quotes, order books, and OHLCV candles.
When building trading systems, analytics workflows, or historical research pipelines, understanding availability time helps users distinguish between market event timestamps and the moment data can be queried or consumed. This is especially useful when working with OHLCV finality, late-arriving market data, and historical backfills.