A setup that performs well during the London session can perform noticeably worse during Asia — not because the setup's logic changed, but because the market conditions the setup was designed around simply aren't present in the same way during every session. Session isn't a scheduling detail. It's a market-condition variable as real as volatility or trend strength, and treating it as an afterthought in your journal means missing one of the more reliable, checkable patterns available in your own trading data.
What actually changes, session to session
The Asia session is typically the quietest of the three — lower volume, tighter ranges, and a tendency toward consolidation rather than strong directional moves, particularly on major pairs without a specific Asia-relevant catalyst. Setups that depend on range compression and mean-reversion (an "Asia Mid Range" style setup, for instance) tend to find their best conditions here specifically because the session's lower volatility is the environment the setup is built around, not despite it.
London, by contrast, is where a large share of daily volume and volatility typically concentrates, especially in the first few hours after the open — a session prone to sharp initial moves as the market re-prices after Asia's relative quiet, frequently including the kind of liquidity sweeps and structural breaks that reversal-style setups are specifically built to catch. New York overlaps with the tail of London for several hours (often the highest-volume window of the entire day) before continuing independently through the U.S. afternoon, carrying its own tendency toward trend continuation or reversal around U.S. economic data releases specifically.
Why a setup can be "right" and still fail on the wrong session
This is the core reason session-blind trading produces inconsistent results even when the setup's entry logic is sound: a reversal setup built around liquidity sweeps needs the kind of volatility London and New York's open reliably provide — take the identical setup into a quiet Asia range and there may simply not be enough real movement for the setup's premise to apply, no matter how correctly the entry rules are followed. The setup didn't fail. The session didn't provide the conditions the setup depends on.
Why session has to be a logged, filterable field
This is exactly why mysetuplap treats session as an explicit field on every live and backtest trade — ASIA, LONDON, NEW_YORK, SYDNEY, OTHER — rather than something implied by a timestamp a trader would have to manually cross-reference against session hours after the fact. A setup's win rate and expectancy can be filtered by session directly in Analytics, which is what actually makes "this setup works in London but not Asia" a checkable fact instead of a hunch a trader half-remembers from a few recent trades.
The overlap windows deserve their own attention
Session boundaries aren't sharp lines — the London/New York overlap in particular behaves differently from either session trading alone, typically carrying the highest liquidity and some of the sharpest moves of the entire day. A trader whose journal only tracks "London" and "New York" as separate buckets, without isolating the overlap window specifically, may be missing the fact that their best results cluster in that overlap specifically, not in either session broadly — a genuinely different, more actionable finding than "I trade London well."
What to actually check in your own data
If you're trading the same setup across multiple sessions without distinguishing them in your journal, the useful next step is straightforward: filter your logged trades by session and compare win rate and expectancy across Asia, London and New York separately, not blended. A setup with a strong overall record can easily be masking a session-specific pattern — excellent in one session, mediocre or negative in another — that's invisible in the blended number but obvious the moment the data is split. Once that pattern is visible, the fix isn't complicated: keep trading the setup during the session where it actually works, and stop trading it, or size down meaningfully, during the session where the data says it doesn't.
