Time zone mismatches in data: the quiet backtest killer

[Amanhasnoname]

Downloaded tick or minute data often comes in UTC. The broker's chart uses a different offset. Mixing them silently corrupts everything built on time.

A strategy that trades the London open will fire at the wrong time if the data is shifted by two hours. A session filter based on server time will run an hour early or late after daylight saving changes. Weekly candles can start on the wrong day if the rollover time doesn't match.

Test results look plausible, because nothing crashes. The strategy just gets applied to conditions it was never meant for.

The remedy is boring. Convert everything to a single reference, usually UTC, handle daylight saving explicitly, and check by looking at a known event. For example, find the NFP release in the data and confirm the volatility spike appears at 8:30 New York time.

A ten-minute sanity check like that can prevent weeks of wasted work.

How do you verify the time alignment of your historical data?