Jason Siegendorf broke in to the trading industry in a non-traditional way, taking advantage of a captive audience of smoking traders.

More than 25 years later, Siegendorf is still thinking differently, moving early on trading innovations pertaining to TCA, automation, broker scorecards, and trading platforms.

Siegendorf, Head of Trading Analytics and Trader at $106 billion asset manager Harris | Oakmark, won Excellence in Trading at Markets Media Group’s 2026 Global Markets Choice Awards.

Traders Magazine, a Markets Media publication, caught up with Siegendorf to learn more.

Talk a bit about your career background, and up to and including your current role responsibilities at Harris | Oakmark.

I started out in prop trading at the CBOE back in 1999. I honestly wasn’t even sure how you got a job there, so I stood outside the exchange and handed my resume to traders on their cigarette breaks.

That got me in as a clerk for Cornerstone Trading. I started on the floor, but I had an engineering background and electronic trading was just starting to take off, so I was able to add more value upstairs, off the floor, helping the firm transition to electronic trading. I wore two hats – trading and technology – and that has stayed with me throughout my career. 

At Harris | Oakmark my responsibilities include working with brokers, designing our custom execution algorithms, managing our automated trading and algo wheel framework, and overseeing our execution analytics. All of this is about optimizing our trading outcomes.

There are seven people in our trading department. Our head of trading and one other trader focus on the Americas, and we have three international traders. I focus on automated trading and algo wheels across all regions – the US, Europe, and APAC – and I still trade sometimes too. I have a junior analytics person reporting to me. The analytics and strategy functions are embedded in the desk – we all sit together and we give traders insight and feedback effectively in real time. It’s a very good iterative feedback cycle.

How has traders’ need for analytics and strategy support evolved? 

Early on traders looked at TCA to just measure how they performed against the benchmark. Now they’re using TCA to ask why they performed the way they did against that benchmark. The actionable insight is in tying specific analytics to specific trading behaviors – traders can use that to adjust their style, they become more aware of their biases, and they make better decisions. So that’s really how we’ve gone about making analytics more approachable and usable on the desk.

How have trading analytics and trading strategy evolved?

When I started out, TCA was mostly basic templates run by our brokers. The data was grouped by things like order size, market cap, and what algo was used. The problem was that it was hard to put context around how that data actually fit our trading – every broker used different metrics and different formats, which made it very difficult to compare brokers. 

A big shift was the buy side starting to build TCA in-house, or having a third- party provider support an in-house platform. We did that, which allowed us to standardize the metrics and organize the data to match how we trade. So we were able to extract more relevant information from the data.

What are the biggest challenges, or ‘pain points’ of buy-side traders? 

One challenge is the signal-to-noise ratio. There’s so much data available nowadays, and the challenge is turning it into something you can act on in the moment. You can have all these dashboards up with all this information coming at you, but it’s still difficult to know what to do differently on an order-by-order basis. We’re very mindful of how we can distill that information down and tie it back to easily consumable and more actionable information. 

Another challenge is liquidity fragmentation. There’s a balancing act of finding natural liquidity and minimizing information leakage, and that’s getting harder because there are so many different venues, and there’s a host of different counterparties with all kinds of different objectives. You want to find natural liquidity, but everybody’s chasing that same natural liquidity, and then you have to protect your intent while you source it. It’s also about knowing where you want to trade and where you don’t want to trade – which venues have mechanisms built in to minimize information leakage, and which ones amplify information leakage. Understanding where you’re trading is very important.

Measurement is its own pain point. We have built sophisticated in-house analytics, but we have to balance how fast we want to answer questions with how confident we are in the answer, so we need to make sure we’re measuring things in a reliable, statistically meaningful way. Data alone is not sufficient – statistically significant data is what’s important.

What initiatives has your team recently completed, or do you have underway?

We’re working on our ability to respond to actionable IOIs in an automated way. 

We like to interact with IOIs by getting a full parent order fill at mid – which is essentially zero slippage. Every automated interaction then becomes a data point we can measure. So over time, we learn which sources of IOI liquidity are worth engaging with and which aren’t, when it is beneficial to interact with IOIs, and when it isn’t. We’re adding brokers as they get the capability to send actionable IOIs into our EMS. 

Another initiative is giving traders a single, distilled indicator of market conditions. For instance, is the market more likely to be trending or mean reverting? That provides a starting point for how to approach a trade, and which tactics and algorithms to use. As the order unfolds and market conditions evolve, they can depart from that, but it turns a lot of noisy information about price action and volatility into one signal the traders can act on with an execution algorithm.

Which industry innovations are improving liquidity sourcing and trading efficiency? Which innovations are not especially helpful or may have unintended consequences?

You’re starting to see a reframing of the block as the holy grail of trading, and more and more you can find equivalent liquidity where information leakage is minimized. Trajectory crosses are a great example. They started as an innovation with a few brokers but they’ve been around for a while now and are widespread. If you trade in a trajectory long enough, it becomes the functional equivalent of a block, but your price risk is spread out over a greater period of time. I see a similar benefit from new platforms like JumpStart and OptimX, which enable traders to find that sub-block liquidity outside of the traditional means of a committed order to a broker. This cuts down on opportunity cost – you don’t necessarily have to guess the right broker algorithm, rather you essentially have an agent out there looking for natural algorithmic flow to cross with.

On the unintended consequences side, principal liquidity is on the rise. Interacting with it can be helpful, but interacting with it indiscriminately is where you run into trouble, because you can leak information without realizing it. A trader needs to understand the business model of the liquidity provider, what kind of orders are suitable to trade, and how to interact with these counterparties to get the best experience.

What are you doing with AI?

Our main use case right now is to generate market notes for the portfolio managers. Traders can spend a lot of time pulling together information from different sources into a consolidated email, and AI can do a lot of that legwork, allowing the trader to go from building everything from scratch to becoming more of an editor of content.

That said, I don’t see AI replacing traders anytime soon. I think there are just too many trading factors that are hard to feed as input into an AI system. Think things like the preferences of the portfolio manager, the differences between broker algorithms that the trader has understood after many uses, pieces of order flow information that come in through different sources over the phone and in chat. All of those are really hard for AI to consolidate right now, and traders do it best.

The real opportunity is in decision support. AI can distill all kinds of information very quickly, it can isolate signals that matter, and it can put historical context around patterns. So ultimately an AI system can be sort of a co-pilot as a decision support tool that can really enhance the aggregation of information.

What are you following most coming out of the SEC?

The SEC is certainly being proactive about challenging regulations that have been in place for many years, asking if they are still necessary. I think there are meaningful changes coming up on the horizon, such as Rules 611 and 610 looking like they’re going to be repealed.

These are changes we need to think about now, even if they might not be in effect for a while. How would a repeal of OPR redefine execution quality? How would we measure best execution differently? These are not bad things; they’re just things that we need to be on top of. 

I think there’s a lot of industry support for what the SEC is trying to do, partly because of the people leading it. Paul Atkins has held a consistent view on 611 since he was a former commissioner, and Jamie Selway comes from industry and understands the challenges we face. 

This SEC is innovation-forward. For example, they want to enable tokenization, which will be a major evolution once it comes to fruition. Also, 24-hour trading seems inevitable, but we need to know what infrastructure changes are needed, what staffing support is needed, and whether there will be real liquidity. This is an innovation that could work both ways, positive and negative, and it’s interesting that just when we were talking about shorter trading hours with more trading compressing around the close, now we’re going in the opposite direction.

What career achievement(s) are you most proud of?

Generally, I’m proud to have been an early mover on a number of trading innovations. 

In 2017 I was given a mandate to revamp trading, and I started by building an in-house TCA platform to standardize our data. Then in 2018 I started in automating trading – we had algo wheels live in both the US and Europe, when there was no blueprint for that. It took a lot of experimentation to get it right, and our broker partners were very engaged because they wanted to understand how the buy side would use these. 

I also designed a fairly comprehensive broker scorecard, which pulls together practical ways that we use TCA to internally diagnose performance. This gives brokers a level of transparency that they don’t usually get, and it’s very helpful for their understanding of how the buy side looks at TCA to improve performance and execution. 

And our firm was lucky enough to be the buy-side partner on the JumpStart platform with ImperativeX, FactSet, and Jefferies. That initiative is starting to take off and it will be interesting to see where it goes.

Lastly, I’d add that it was an honor to be nominated for Trader of the Year in 2024, and to win the Excellence of Trading award from Markets Media this year. That’s a genuine career highlight for me.