The hour mark: a small clock effect worth checking in your own data
Posted: Sat Oct 31, 2026 8:39 am
Something odd shows up when a trade log is grouped by minute of the hour. Entries in the first few minutes after the hour and in the last few minutes before it often behave differently from entries in the middle. Part of the reason is human: many participants and algorithms schedule actions on the hour, data releases cluster at the half hour and the hour, and fixings and option expiries are keyed to round times in various centres.
That does not mean there is an edge in trading the hour. It means volatility is not spread evenly across the sixty minutes, and a setup tested across all of them may be hiding a good half and a bad half.
An easy exercise: tag every trade with its minute of the hour in a spreadsheet, bucket them into six ten-minute groups, and compare average result and average spread paid. Many traders find one bucket carries the losses, and that the cause is either slippage or choppier price right after a release.
Do not read too much into a small sample. Fifty trades split six ways is thin. Wait for a few hundred before changing any rule, and check that the same bucket shows up on more than one pair.
Which ten-minute window of the hour is the weakest in your own journal, and does it hold across pairs?
That does not mean there is an edge in trading the hour. It means volatility is not spread evenly across the sixty minutes, and a setup tested across all of them may be hiding a good half and a bad half.
An easy exercise: tag every trade with its minute of the hour in a spreadsheet, bucket them into six ten-minute groups, and compare average result and average spread paid. Many traders find one bucket carries the losses, and that the cause is either slippage or choppier price right after a release.
Do not read too much into a small sample. Fifty trades split six ways is thin. Wait for a few hundred before changing any rule, and check that the same bucket shows up on more than one pair.
Which ten-minute window of the hour is the weakest in your own journal, and does it hold across pairs?