Institutional trading desks today operate in an increasingly fragmented and automated market, where execution quality depends on both advanced technology and deep market expertise. At Rosenblatt Securities, a boutique agency-only broker with roots on the floor of the New York Stock Exchange, that balance remains central to the firm’s approach. In an interview with Traders Magazine, Jourdan Frain, Chief Operating Officer at Rosenblatt Securities, discusses how the firm structures its trading desk, balances human judgment with automation, and adapts its execution strategies to shifting liquidity patterns and emerging market structure trends.
Please tell us about Rosenblatt’s trading desk. How is it structured and what makes it unique?
Since our founding 46 years ago, Rosenblatt has been an institutional equities, agency-only broker. The firm started as a $2 broker on the NYSE floor, added different business lines and trading desks through the years and now our Execution Services unit is comprised of not only the largest NYSE floor broker, but also portfolio, high-touch and low-touch trading. All these desks and traders work together closely, functioning as one team to give clients the access and feel that best suits their trading needs. Together, they traded more than 32 billion shares last year and more than $2 trillion notional.
It’s somewhat unique for all of those trading modalities to be offered natively, particularly from a boutique. And that boutique environment also provides a level of coordination among the different trading desks that presents as one face to clients. That extends across the firm as traders work hand-in-hand with our preeminent market structure and in-house development teams to give clients a holistic and largely bespoke offering.
How do trading volumes influence your execution strategy across different clients?
The basic answer would be the pacing of execution, but that is largely dependent on client intent and the specific characteristics of the order being sent. It also isn’t just total volume, but where the volumes are occurring and how accessible they are, particularly with roughly half of all volume now taking place off exchange and a big chunk of that being retail that goes directly to wholesalers. Because liquidity has grown more concentrated at the close, we’ve devoted a lot of time, energy and personnel to being one of the biggest players on the Street at and near the close. But volume is only one of many inputs, including spreads, impact cost and timing. All these factors, and how they interact with one another, affect what strategies we employ, along with when and how.
Rosenblatt emphasizes execution tailored to each client. How do you balance customization with speed and efficiency?
No client is the same, and much of the customization happens prior to an order getting sent. Knowing your client, working to understand their intent and expectations before the first trade does a lot of the heavy lifting. After the first trade that relationship with the client becomes iterative through data analysis and communication. That naturally becomes more seamless and thereby faster and more efficient.
How has the evolution of electronic trading changed your approach to execution quality?
The market continues to grow more automated, with liquidity fragmented among more venues, order types and strategies. It’s also getting more complex. Fortunately, electronic tools have evolved with the market to help traders manage that complexity. Data acquisition and analysis has increasingly become a necessity, and that comes into play at all levels. Our team is analyzing not only the performance of a portfolio or an order, but also looking child-level metrics such as spread capture, fill rates and hit rates, including at the venue level. This is particularly important when comparing competing order-routing models and venues. Rosenblatt’s execution team, in collaboration with our developers, have helped deliver some of this analysis and data to clients. Aside from working with clients on customized TCA, last year we relaunched a web portal and consumable API feed called Early Imbalance Tracker, which delivers to clients the early closing imbalance data the NYSE disseminates through floor brokers. That helps our customers make better decisions on their desks.
What metrics do you prioritize when measuring execution success?
As an agent our first priority is to get the best execution price for a client, with performance measured against each client’s individual benchmark. Subjectively, our key metric is client satisfaction. That said, there is a lot of objective analysis using a variety of metrics to get to that point.
The team is in continuous communication about changes to performance, market conditions and strategies, and we formally review our execution data monthly. We meet quarterly to review the overall performance of our routes and strategies, internal and external, propose any changes to them and generally discuss any themes that are occurring with our clients or our execution partners. We evaluate the aggregated performance of individual algorithmic strategies against their respective benchmarks, such as VWAP or implementation shortfall. Rather than focus only on the weighted aggregate numbers, we appraise other statistical measures like the median and variance of order performance. That gives us a clearer view of what to expect when executing with a particular counterparty or strategy. Our market impact model also gives a solid estimation of how the order did based on its characteristics. Finally, when evaluating algos in our own suite, Q-Edge, we are looking at reversion markouts at the child level to quantify the clear advantage we’re delivering for our clients.
How do you maintain an edge in liquidity sourcing across different market conditions?
Given the wording of this question, it’s impossible not to lead with our algo offering, Q-Edge. Late last year, as a result of NYSE’s continued systems improvements, we were able to seamlessly integrate floor-broker order advantages into our strategies. That had never been possible before, and it helps clients in two ways. First, it makes participating in the closing auction, with maximum flexibility, even more efficient. Second, and most important, we can finally embed into an algorithmic strategy our floor-brokers advantage on resting limit orders. Unlike most US exchanges, the NYSE does not use pure price-time priority to determine how such orders are filled. We can use e-Quotes on the NYSE to achieve fill parity at a price level even if we’re not first in time priority. Liquidity-seeking interest is instead allocated roughly equally against floor broker e-Quotes, the Designated Market Maker and the first order on the book at the best price. Clients always have been able to access e-Quotes through floor brokers, but NYSE systems previously could not accommodate making them available via algorithms. The parity advantage significantly improves fill rates, reduces adverse selection and reduces transaction costs for many passive strategies.
Being a pure agent is also, in some ways paradoxically, inherently an advantage to finding liquidity. We are unconflicted as to where and how to go to source liquidity for our clients—we don’t have, nor do we want, a proprietary dark pool that might be a tempting destination for fee avoidance. One example is the rise of bespoke IOI feeds from market makers. They can look at our order profile, incorporate that into their model and potentially provide an improved market.
What’s the biggest challenge when managing a trading desk focused on highly individualized client strategies?
Individually our traders have outstanding knowledge of and relationships with their clients. They combine that with an intimate knowledge of our tool set and can lean on in-house resources like our market-structure and software-development teams to provide even more individualized service for clients. The challenge can be keeping that cohesive across the firm and not falling into the easy trap of becoming too siloed; ensuring that the sum is greater than the parts. I came to this business from the military, so I go back to the tenets of teamwork, intelligence sharing, everyone understanding the overall mission, and cohesion all being critical to success. For me these remain applicable for Rosenblatt’s trading desks, which need to operate as one unit.
How do you decide when to rely on algorithms versus human judgment?
In today’s environment, it’s critical to rely on both humans and technology simultaneously. They are always working hand-in-hand. A trader executing a high-touch order that gives us discretion, for example, needs to quickly synthesize market conditions, the order’s characteristics, the client’s benchmark and the available execution methods to apply the best strategy with the appropriate parameters. This requires not only extensive market knowledge and a deep rapport with the client, but also complete command of every venue, order type, algorithm or other tool at their disposal. Even on orders directed to specific venues or sent to algo wheels, our traders have to apply that knowledge while monitoring progress and potentially changing course as the situation dictates. That is equally true on the NYSE floor, where our brokers operate on a largely electronic exchange platform.
Looking ahead, what trends in trading or market structure are shaping how Rosenblatt approaches client execution?
I’ll first go back to the NYSE’s modernization of floor-broker systems. It lets us extend our unique capabilities and the improvements to execution quality associated with them to a wider range of clients. This is part of a broader trend in which venues, brokers and clients are seeking alternatives to the price-time priority rat race that de-emphasize speed and deliver better execution quality. Other manifestations of that trend include increased use of innovative ATSs like IntelligentCross, PureStream and OneChronos — all of which center on alternatives to price-time priority matching — as well as a move toward bilateral and highly segmented trading in ATS private rooms, like LeveL, and single-dealer platforms. We of course continue to seek blocks for our institutional clients given their liquidity demands, unfortunately block market share continues its long-term decline. In 2025 only 5.44% of market-wide volume was executed in block size, the lowest level since our market-structure team started tracking that statistic in 2014. We do worry a bit about an ever-growing share of total trading volume moving toward bilateral or highly segmented, off-exchange models, and what impact that might have on price discovery. But to the extent clients demand to trade that way, and it improves outcomes, we have an obligation to participate. This brings me back to our Q-Edge algos, which give clients another option for de-emphasizing speed, reducing adverse-selection risk and improving execution quality — all while still trading on the NYSE, the single biggest source of Tape A market share and an underutilized destination for Tape B and C names. Taking advantage of floor-broker parity improves fill rates and mark-outs for passive orders, which shows in the data. Continued use of d-Orders on the close gives the buy side more flexibility to get the closing price without locking in to the MOC too early and missing the opportunity to trade at potentially better prices while liquidity builds into the closing print.
Of course we’re also watching closely potential regulatory changes, including the prospect of a repeal or rollback of Rule 611 of Reg NMS (the Order Protection Rule). But like most other brokers, we’ll continue to seek the best possible prices for client orders wherever they reside, regardless of whether Rule 611 exists in its current form or is removed or watered down. The wider reconsideration of Rule 611 might also affect whether already-adopted changes to tick sizes and the Rule 610 cap on exchange access fees take effect as scheduled in November or are delayed or canceled. But regardless of the outcome we’ll be ready and continue with the same approach to superior execution quality we’ve always pursued.
