Diagnosing Delays in Live Five-Minute Indicator Updates

Summary

The post asks why a five-minute technical indicator in live trading appears to update three to four seconds after a period ends, even though historical testing updates it at the period boundary. The author reports connecting to OKEX futures from a rented Alibaba Cloud server in southern China and measuring the records-fetch call at under 100 milliseconds. The question is whether the lag comes from the platform, exchange, or another part of the data path.

This is a troubleshooting report rather than a confirmed diagnosis. Its useful distinction is between the time spent fetching records and the later availability of a completed candle’s value: a fast API response does not establish that the exchange has finalized and published the new bar. The post gives no answer, comparison across providers, timestamp analysis, or resolution, so it cannot identify the source of the delay or show that the reported behavior is typical.

Key ideas

• A live five-minute indicator may show the prior period’s value several seconds after the period ends.
• The author measured record retrieval below 100 milliseconds, which does not by itself explain when a completed bar becomes available.
• The post raises a latency diagnosis question but provides no confirmed cause or fix.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.