With regulators placing an ever-greater focus on best execution, firms across the buy and sell side are making increasing use of transaction cost analysis (TCA) to demonstrate to customers how their execution performance measures up against various benchmarks. But TCA is not just a regulatory tool. Real-time TCA can also be an incredibly useful resource on the trading desk, providing valuable – and actionable – insights to traders.
However, challenges do exist. Effective TCA relies on multiple sources of data, which can be spread across various systems, often disconnected from one another. And in order to generate useful insights, TCA needs to take a range of data points into account, including historic execution performance, current market prices, and the status of orders being worked.
In the US equities markets, the challenges are further exacerbated by the fact that the markets are highly fragmented, which can make it difficult to identify the best available prices across lit and dark venues. Also, once an order goes to a broker, the execution path through the broker’s algo suite can be complex and opaque.
Toronto-based software company Spacetime.io aims to address challenges such as these with its TradeFabric Intelligence platform, which at its core is an advanced, real-time TCA engine.
“Fundamentally, there has been an explosion of information available to traders with the increased number of venues, higher pace of electronic trading, increasing use of algos, and the resulting data rates,” Spacetime.io’s co-Founder and CEO Evan Young tells TradingTech Insight. “At the same time, trading has become less hands-on, with a ‘fire and forget’ aspect to algos, leading to less visibility for traders on what is happening during the trade. It’s important for people to regain that visibility and control.”
Young explains that the company’s innovative approach to TCA revolves around three key pillars: making it more visual and intuitive; more timely; and more relevant.
“From a visualization perspective, TCA generates lots of numbers, which can make finding significant signals and outliers challenging. Having visual ways to see and understand data is therefore critical,” he says. “Regarding timeliness, analysing trades at the end of the day, month, or quarter is certainly important, but it doesn’t allow for on-the-fly fixes to correct course while the trade is still ongoing. Generating alerts and insights that are relevant and adaptive to changing market conditions is where we go beyond standard TCA. We strive to enhance decision-making by pulling all the necessary information together, helping to visualize it, and surfacing meaningful insights.”
For the buy side, this approach is invaluable, says Young, particularly in today’s low-touch electronic markets. “Brokers aren’t able to give you the same level of market colour as they did before, when you were trading with them using care orders,” he suggests. “Now that you’re using the electronic trading tools yourself, there’s a gap there that needs to be filled, answering questions such as ‘what is really happening in the market? What should I do with this order? Where is liquidity going off? What’s happening in the dark?’ All the things that my broker used to call me on. We want to give institutional traders the tools they need to regain the visibility and market feel they used to have, and ultimately to be able to change direction during a trade to deal with potential issues or react to market opportunities.”
Andrew Karsgaard, Spacetime.io’s Director, Product Management, gives an overview of what TradeFabric does. “We focus on understanding the performance of large institutional orders by examining how they interact with the market over their lifetime,” he says. “We analyse the state of the market at the time each fill occurred. We know the exact location of each fill, which broker executed it, which algo it came through, and which venue it executed on. We perform granular analysis on things like adverse selection or spread cross at a venue level. From there, we can aggregate fills to assess the performance of brokers and algos, including the scheduling and routing aspects.”
He continues: “Traditional TCA focusses on what has happened historically, but we wanted to provide insights throughout the order lifecycle. We have a pre-trade model that we use for single stocks and baskets to decide how best to execute. The goal is to make recommendations based on live TCA so that you can see estimated costs and the optimal type of execution based on different algos, different lengths of time, and various parameters that consider the current trading environment. The aim is to generate a recommendation on the best way to approach a trading situation.”
The execution process can change depending on what happens in the market, giving traders the opportunity to shift from one algo to another, explains Karsgaard. “It’s about identifying significant changes in the market while we are trading,” he says. “It may well be that there’s a huge amount of liquidity in the dark today for example, and traders are going to want to know that. Then, given that they know it’s happening, they can change their order to react to those new conditions: become more or less aggressive, or change where an order is routed.”
TradeFabric is also targeted at sales traders on the sell side who are dealing with providing the same level of service, often with smaller desks and a more electronic client base, says Young. “If there’s an issue, you want to call your client before they call you. Or if there’s an opportunity, you want to be alerted so you can make that outbound call. And if you’re a broker using other brokers to execute downstream, you often have very little visibility into how those orders are being handled, so you face similar issues to the buy side in terms of understanding what’s going on and being able to manage those downstream brokers and algos.”
A key aspect of TradeFabric is its open architecture, which facilitates easier integration with OMS and EMS platforms, says Karsgaard, who believes that vendors in the O/EMS space are becoming more receptive to the idea of using APIs, FDC3 at the desktop level, and other integration methods behind the scenes, thus becoming a central backbone for workflow where other elements are plugged in for analytics, risk, client reporting, or whatever else might be required.
“This approach is much more prevalent now,” he says. “The industry is moving towards a more open, microservices-based architecture, which benefits the trader because you can knit these things together to create integrated workflows that incorporate analytics, risk, trading, order management, and pre-trade compliance, without having to swivel in your chair every time you move from one task to another. People don’t want more things on their screen if they can avoid it. So the more integrated we can be, the better. And underlying it all is the fact that we’re cloud-based, but we’re also API-based. All the front-ends within the system use our own APIs, and we can expose those APIs to clients. So, if clients or O/EMS vendors have that ability, they can work with the data that’s available. And that’s becoming increasingly common.”
Subscribe to our newsletter
