Not all hours are equal. This experiment tested whether restricting entries to the highest-volume window improves results — and it did, meaningfully, by trading less.
Take the same setups, but only during the London–New York overlap (07:00–15:00 GMT) when gold volume and directional conviction are highest. Everything else about the strategy is unchanged.
Trade frequency dropped 38%, but profit factor rose from 1.22 to 1.35 and max drawdown fell to -3.1%. 26 of 32 folds passed. Fewer, better trades — the recurring theme of the whole course.
Outside the overlap, gold often drifts in thin, mean-reverting chop that punishes breakout and sweep entries. The overlap concentrates the real institutional flow, so the same signal has a much higher hit rate when it's confirmed by volume context.
Volume context is non-trivial. A filter that removes a third of your trades and improves every metric is one of the best trades you can make — because the trades it removes are the ones with negative expectancy.