Direct answer
USD concentration is most “active” when trading sessions overlap and overall liquidity is high—typically when London and New York hours intersect, because more market participants are active at the same time. That does not mean USD concentration is always strongest at the same clock hours worldwide; it depends on when the relevant buy/sell pressure for USD is strongest in the specific market environment.
What USD concentration means (mechanism)
“USD concentration” is an informal way to describe how strongly USD tends to dominate currency pricing and risk exposure relative to other currencies. In practice, you can think of it as the combined effect of:
- Positioning and funding demand: how much trading and hedging activity involves USD.
- Cross-currency correlations: how other currency pairs tend to move together when USD moves.
- Liquidity and order flow: how easily traders can buy or sell USD at different times.
A key point is that USD concentration is not a single universally published metric. It is usually inferred from market behavior—such as whether USD moves explain a larger share of variation in other pairs—rather than directly observed as one number.
How session overlap changes intensity (non-real-time model)
Forex liquidity and trading activity tend to rise during major global session overlaps. A simplified, non-real-time way to reason about this is:
- More participants active → deeper books: when two large centers are open at once, there are more buyers and sellers and tighter typical spreads.
- More simultaneous hedging and execution: traders adjust exposures across currencies, which can increase the “USD driver” effect.
- Faster adjustment of correlated pairs: if many pairs react to USD moves, you may see stronger USD-linked behavior.
In that framework, USD concentration is often most noticeable during the overlap of the two largest trading windows (commonly described as the London–New York overlap). Outside overlap, liquidity is usually thinner, price discovery can be slower, and USD-linked effects can appear weaker or more erratic.
A worked example (assumptions stated)
Assume you compare two equal-length windows in a generic week:
- Window A: a major overlap period.
- Window B: a period with only one major session.
If, during Window A, USD-related moves explain more variation in other cross-currency rates and bid/ask spreads are typically narrower, then USD concentration would be observed as “more active” in Window A by your chosen definition. This is an example of a verification method, not a guarantee of results.
Limitations and failure modes
USD concentration can be misleading if you treat it as stable or time-fixed. Common limitations include:
- Market conditions override session effects: events like sudden risk changes can dominate over any typical liquidity pattern.
- Costs and execution differ: wider spreads or slower fills in off-overlap hours can change observed behavior even if underlying pressure is similar.
- Different ways to measure give different answers: “most active” depends on whether you define it using correlations, relative volatility, or sensitivity to USD.
- Historical relationships may not persist: the fact that USD-linked behavior appeared stronger in past overlaps does not ensure it will do so in the future.
Verification and next question to answer independently
To verify which session is most relevant for your own definition of USD concentration, compare the same kind of evidence across session windows (for the same duration):
- USD-linked sensitivity: how much of other pairs’ movement is associated with USD moves (using your chosen statistical approach).
- Liquidity proxies: whether spreads or realized volatility are consistently different during overlaps.
- Robustness checks: repeat across multiple weeks and avoid mixing different event regimes.
Next, define your exact measurement rule for USD concentration (e.g., correlation-based, variance-explained, or spread-adjusted sensitivity). Once that rule is explicit, you can map “most active” to the session windows that score highest by that rule—without assuming a universal answer.