THETRADETECHFX DAILY from the floor
How do you distinguish genuine liquidity from liquidity that simply looks good on screen? It is a generally accepted reality that often, displayed liquidity and executable liquidity are two different things. A tight price on a screen is the starting point. The real test is whether that liquidity survives contact with your order. Does the provider maintain their pricing as order variables such as size increase? How consistently do they fill? How often do they reject or requote? And most importantly, how do they behave when volatility picks up? I think the industry has become much better at measuring this than it was even a few years ago.
Rather than relying on anecdotal experience, we can now assess providers over thousands of observations, looking at hit ratios, quote quality, response times, fill consistency, and execution outcomes across different market conditions.
One of the biggest mistakes is judging liquidity purely by the current best price available. Execution quality is really about situating yourself in a place where optimal execution is a probability rather than a possibility. A provider who is often secondbest but reliably executes with consistency can deliver better long-term outcomes than one who sporadically tops the price ladder but disappears when you actually need them.
Are traders at risk of drowning in execution data rather than generating useful intelligence? No, I don’ t think the problem is too much data. Most execution desks now have access to more analytics than ever before. The challenge for desks now is to make sure they ask questions of the data, the right questions, and ensure that outcomes are questioned and if required, behaviours are changed.
An analysis that doesn’ t alter how an order is executed, how a trade is sized, how a provider is selected, or how you engage with liquidity, is at best interesting intelligence – but it’ s probably not useful. At worse, it’ s noise. Context matters. Looking at a single execution metric in isolation is not helpful, and good analytics move away from being just an autopsy. We’ re also moving towards a world where traders spend less time gathering information and more time interpreting it. That makes judgement even more valuable. Data should support experience, not replace it.
How much has analytics changed your routing decisions over the last two years? Our decision-making has evolved, but not in a‘ black box and turn it on sort of way’, it’ s a continuous feedback loop with human
Finding genuine FX liquidity
The TRADE sits down with JOSEPH FORDE, FX trader at Brown Brothers Harriman, to unpack the FX liquidity landscape and how analytics is increasingly enabling traders to make dynamic routing decisions while retaining human judgement.
input at each stage.
Historically in FX, routing decisions relied heavily on trader experience and established relationships. It’ s important to stress that those remain incredibly valuable and always will be, they are foundational to our asset class corner of the world. But analytics now allow us to validate and robustly challenge that intuition. It has allowed us to shift from static counterparty lists to something more akin to dynamic conditional routing within a set list of counterparties.
Today it’ s possible to evaluate providers across multiple dimensions. We’ re looking at execution consistency, behaviour during volatile markets, fill quality, market impact, and how different providers perform under different trading conditions. One area that’ s become particularly important is recognising that there is never a universally
“ best” liquidity provider. Performance varies by currency pair, order size, time of day, and prevailing market conditions.
The most effective execution combines quantitative evidence with human judgement. The biggest practical change in the industry has been speed: what used to be a quarterly TCA review is now closer to a standing input into how to approach an order today.
Which execution metric is most commonly misunderstood by the buy-side? If I had to pick one, it would probably be average spread or average execution cost when viewed in isolation. Aggregate averages are useful, but they can hide what is actually driving performance. Two providers might produce the same average outcome or show an average spread while behaving very differently across market regimes, trade sizes, or liquidity conditions.
I also think there’ s a tendency to focus on the outcome of single large individual trade rather than the consistency of the execution process. Any provider can have a great day, what matters is whether they repeatedly deliver good outcomes across thousands of executions. This is where more granular analysis becomes valuable.
Ultimately, no single metric tells the whole story. Good execution is multidimensional. It’ s about balancing price, certainty of execution, information leakage, market impact, and consistency. The most optimal execution frameworks don’ t search for one holy grail number – they combine multiple measures to understand the complete execution process.
20 THETRADETECHFX DAILY