Direct answer
Pair spreads are usually most active—meaning they change more noticeably—during the time windows when trading sessions overlap and market liquidity is highest but also uneven across pairs. For many forex hours, this often involves the overlap period between the London session and the New York session. In contrast, during quieter hours with fewer active participants, pair spreads can be wider and more erratic because fewer orders are available at each price level.
Because spreads depend on conditions, you should treat “most active” as a pattern of typical liquidity behavior rather than a fixed schedule that always applies. Also note that different brokers and venues can show different spread observations, even for the same underlying market.
Mechanism or definition
A “pair spread” is the difference between the bid and the ask for a given currency pair. It is often expressed in pips. The spread reflects how much it costs to enter immediately at the ask and exit immediately at the bid, given available buy and sell interest.
Two stable mechanics matter when comparing sessions:
- Order book depth and liquidity: When more participants trade, there are usually more bids and asks close together, which can reduce the spread and make it change more frequently as demand shifts.
- Volatility and risk control: When markets move quickly, providers may widen spreads to manage inventory and execution risk. Even if liquidity is “higher,” rapid price changes can still cause spread expansion.
Session overlap increases the chance that both mechanics apply at the same time: trading activity rises, but so can volatility driven by different participant groups and time-zone news flows.
Evidence or example (non-real-time)
A simple non-real-time way to reason about timing is to consider three periods: (a) a major session that has strong participation, (b) a second major session starting later in the day, and (c) the overlap.
Assume you observe a currency pair whose spread is measured continuously. In a typical daily pattern:
- Outside peak hours (fewer participants): bid–ask depth is often thinner. A small shift in order flow can create a larger jump in the nearest available bid or ask, so spreads can appear wider and more “spiky.”
- During overlap (more participants from both regions): you may see more frequent changes because more orders arrive and are canceled at different times. Spreads may tighten during calmer moments, but they can also widen briefly around sudden bursts of demand or volatility.
- After overlap (one session fading): liquidity can gradually thin again. Spreads may remain variable, but the amplitude often changes.
This reasoning helps explain why people often report the biggest spread activity during London–New York overlap, without claiming a precise minute-by-minute rule.
Limitations and risks
Several limitations can prevent “session-based spread activity” from being a reliable standalone guide:
- No real-time guarantee: The market can deviate from typical liquidity patterns. A calmer day, a major risk event, or a temporary technical issue can shift spread behavior.
- Provider and execution differences: What you observe as a spread can vary by venue, broker model, pricing feed, and execution method. The same session may look different depending on how quotes are formed.
- Costs and measurement effects: Some trading platforms display raw spread while others incorporate additional components (for example, commission-like costs). Two traders can compute “effective spread” differently.
- Failure mode: overfitting to history: Even if spreads moved a lot during a certain overlap on past days, historical relationships do not establish future results. You can also mistake short-term noise for a stable “session signal.”
Verification or next question
To independently verify the timing for a specific pair and data source, you can compare spread behavior across session windows using your own historical quote data (for example, computing average and range of bid–ask differences by hour-of-day). Treat the result as specific to your chosen data feed and instrument.
A useful next question is: “Which component is driving the spread changes for my data source—liquidity depth, volatility, or quote formation?” If you can separate these, you can better interpret why spreads are more active in some hours than others.