088: Dave Lauer – A Former-High Frequency Trader Talks Speed, Market Structure and Dark Pools

Chat with Traders is brought to you by Trade the Pool. Did you know that every decade the market reinvents itself? Online brokers opened the doors. Mobile apps made trading seamless. and commission free trading erased barriers. Now a new era has begun. Meet, trade the pool, limited risk trading. And now you also have unlimited time to reach the profit target. From now on, your trading risk is capped. and your trading opportunities are limitless.

Trade the pool funds home-based stock traders with up to$200,000 in buying power. That means you can trade larger positions and scale your strategies without risking your own savings. It's time to trade with more capital, making it truly worth your time and effort. Ready to trade the pool? Click the link in the description and join the stock trading revolution today.

Are you ready to get serious about trading? Then join Tasty Trade, Investopedia's best platform for options trading in twenty twenty six. Options, futures, and more. Tasty Trade has everything you trade all in one platform. Get low commissions, including zero commissions on stocks. So you can keep more of what you earn. Trade smarter with advanced charting tools, a pre-built strategy selector, risk analysis tools, and more features. Visit Tastytrade.com slash.

Chat for more information. Tasty Trade Inc. is a registered broker dealer and member of FINRA, NFA, and SIPC. Podcast. What's going on team? Welcome back to another episode of the Chatwith Traders podcast. I know I say it all the time, but thank you so much for being here. I really appreciate you tuning in. My guest this week is Dave Lauer. Dave is a former high frequency trader for firms such as Citadel and Austin Trading.

He's worked specifically in various areas of the HFT pipeline, including research and modeling, building hardware, and programming and operating strategies which are measured in millionths of a second or microseconds. Following the flash crash, Dave left his role as a trader for various reasons we discussed during this episode, and now as a partner of Core Group, he consults to institutional managers on market structure and best execution.

Dave was also featured in the V Pro documentary, The Wall Street Code, along with other chat with traders guests, Haim Bodek, Eric Hunseder, and Blair Hull. I really enjoyed speaking with Dave and I hope you'll enjoy listening just as much. Show notes for this episode can be found at chatwithraders.com forward slash eighty eight. Now here it is, my interview with Dave Lauer.

Start by telling us a little bit about how you got your start in financial markets. Like where did all of this begin for you? Um yeah, I'll I'll try and keep it quick, although I I like to tell the story. I think it's kind of a funny story, but

I got my start as a kid. So when when I was young, uh and I was playing around with a little thing, you know, called AOL, which was America Online, um And you could put together this sort of uh paper I guess we would call it now like a paper portfolio of paper trading. And I just put all these companies that I knew and it was the mid nineties and so the companies I knew were Intel and Microsoft and Cirrus Logic and US Robotics and that kind of thing and

And I said, Oh, look at this. I could buy this portfolio for two thousand dollars. Um and so I told my parents about that and they just kind of dismissed it. I think I was like, you know, 14 years old or something. And then I, you know, like

Uh a year or a few months later it was like, Oh, hey, look, this portfolio's worth four thousand dollars. I hey my pa I went to my parents, I'm like, You really should buy this and they just said, Oh, you're you don't know what you're talking about, you know, forget about it. And a year later, you know, it had doubled again and it just seemed to keep doing that. And so um eventually we started a stock club and

uh focused on high tech and that was a pretty lucrative thing to do in the mid late nineties, uh going into two thousand. And so that's kind of where I I got the bug for uh markets and trading. And so when I went to school, I did computer science and I got a master's in finance. And I came out and I I still wanted to do high-tech. So I started at a startup that was funded by Goldman Sachs.

And we built a piece of hardware for accelerating low latency trading systems and low latency trading desks. And so I did that for a few years um from uh f oh five to oh nine and it was a pretty good time to sort of be involved in that part of the industry.

And then I eventually decided I should be doing the trading, not selling equipment to the trader. So I became a high frequency trader. And I worked at Citadel and then at Alston Trading. I did that for a couple of years and I did everything from quantitative research and modeling to programming the trading strategies.

um and you know, managing them and executing, you know, dealing with them, executing trades, that kind of thing. And and they scaled up pretty large. I specialized in ETF market making. uh auctions, opening and closing auctions, event-based trading, like around news events, um and We also did quite a bit of interlisted arbitrage between different geographic markets like US and Canada, for example. Um and that was that that ran through twenty ten and the flash crash uh where I was working at

One of the largest liquidity providers for the e-mini market. Um, and so obviously the flash crash was. an interesting event to see from that perspective. It was on the desk right next to me and I was really good friends with those guys. Um and it was it was kinda eye opening so I think

Within a year after that I had left high frequency and I'd left financial services and I really had no interest in the industry anymore. I was kind of disgusted with it. Um and I built a storytelling website with a with a good friend of mine. Um and the first story I told was why I left financial services, why I left high frequency trading.

Uh and it again it was twenty twelve so it was kind of fortuitous. It was just a good time to to talk about that kind of thing. And someone from a radio station in the US, National Public Radio, asked me to come on and tell the same story, uh, and I did.

And the next thing I know I was getting calls from the US Senate and the SEC and from Brad Katsuyama at IEX and Um, you know, so I kind of found this new path in back into the industry where I realized I can, you know, kind of take what I had learned And use it in a different way. And so ever since then, that's been what I've done. I've tried to help people understand market structure. I've really focused uh on institutional asset managers.

and helping them reduce the the friction of interacting with markets, reduce the cost, reduce the opportunity to leak information and have slippage, that kind of thing.

Right, right. Well, I really want to break this down because this is super interesting. So just going right back to when you've mentioned uh that you were that you got involved with a startup that was funded by Goldman Sachs. I'm really keen to hear a little bit more about that. Um, you what were you exactly working on there? Was it the actual hardware and and not the sort of the software, but more of the hardware that's used for high frequency trading, is that right?

Yeah, so we built a piece of hardware. Uh the company's still around, um, although they've they've since shifted. Uh but w what we did was we built this this piece of hardware for something called middleware messaging. So, you know, uh at at almost every high frequency trading desk Uh you have all these market data feeds coming in, right? And then that that data has to get sent out to lots of different models. Those models have to act on that data.

and then send orders to the market. You know, and that can that that is true whether you're dealing with, you know, massively distributed co-located systems or not. Um and so W the hardware that we built, we actually put feed handlers in the hardware. Um we'd we'd started with like Itch, uh ARCA, we had an Oprah feed handler, which at the time was something that almost nobody could contend with those data volumes. They were just massive.

Um and so we had actually built this accelerated networking appliance that would go into the data center and you know, like I said, Goldman was one of the funders and so I spent quite a bit of time at Goldman. um, was a contract worker there for over a year and sort of got to play in their development lab and data center, which was just for a tech geek kind of the craziest thing that, you know, you can imagine. It was just a ton of fun.

Right. So had you had much experience with the actual hardware aspect of things uh prior to this? No, no, I'd always been uh I'd always been a software guy, although you know, I've been building my own computers since Well, since we built my first one with my dad when I was seven, but you know, so I I I know I knew some stuff, but I was very new to the hardware space. I was I really came came into it as a programmer. Okay. Sure, sure. So

You know, that that that hardware that you were working on, where did that actually fit into things? Like where was that? Where did that go, I guess, is the question. Like did it go in the in the data centers? Yeah, it would go in like the the main the firm's main data center usually.

um where you would bring all the market data feeds in. Um although some did want it in colo spaces and that was more practical you know, if you were in a colo space that had multiple exchanges, obviously, it was a little expensive to put a pie w a piece of this hardware in every colo space, but eventually, not not at the company I was at, but others.

went that route where they started putting these things in cards rather than in appliances like we did. And then it made a little more sense to go uh, you know, to all the individual colos. Okay. And was this used uh solely by Goldman or was it used industry wide?

Yeah, it was it was industry wide. I mean we had it it was a really interesting way for me to be introduced to the industry. You know, I'd never worked on Wall Street before, never done anything, you know, professional financial services, and being the third employee at at this startup Um, you know, I was in all you know, I was in almost every meeting at at every major company, you know, every major big bank, most of the hedge funds, all the prop shops.

um and you know, understanding kind of how they all worked and what their their problems were. And then how, you know, maybe what we were doing could help with some of those things. Right, right, got it. So tell us a little bit more about um, you know, the next step from there where you went on to Citadel. Um, I'd be keen to hear about the type of things you were employed to do there.

Yeah, so you can imagine I don't really talk specifically about what I did there. Um I can I'm happy to talk kind of generically about my years in HFT, um but I It's still a very secretive industry and not wanting to ever take the risk of litigation. I just generally don't talk specifically about companies that I worked at. Okay. Okay. No, that's cool. That's cool. I totally understand that. Yeah.

Not being a lawyer, I just assume I if I don't do that I'm okay. All right, so we talked a little bit about how HFT firms optimise their strategies for speed. Um, you know, besides the the actual hardware, like the types of hardware that you worked on, are there any other ways that they Like like what are some of the other ways that HFT firms to use to really reduce their latency as much as possible?

Yeah, so it's funny because w you know, working at at Tervella, which was the startup I worked at, um, you know, we solved one part of the problem, but you were it didn't matter if you solve one, you know, block in a pipeline, if you you know, you're only as fast as your slowest point, right? So And that time, in those few years, I became a real um latency measurement and and mitigation specialist. That's really what I focused on.

And I came up with different techniques for measuring it, ones that are sort of just starting to become more mainstream. Um just a couple years ago you saw the Financial F Information Forum adopt one of our techniques.

uh that we had published a paper on which was about instead of looking at um volume in terms of messages per second uh we used to look at it in terms of um messages or network frames per millisecond because the micro bursting uh at the millisecond level was uh a major component of latency and so

you know, you needed to control for that and be able to handle micro bursting on on that timescale. And I I'd imagine these days it might even be, you know, they might they might be down to the microsecond. But You know, we would look for opportunities

to optimize latency that you would measure in double digit microseconds. Um when I had built one of my trading strategies when I was actually in HFT, um and believe it or not this was in Java Uh but it was running on a ser on a server in a colo space and I would measure every message in and out of that box. So it was an event-driven strategy. So it was very easy to know when the message came in that I was reacting to and the order went out.

which was the reaction, and I would measure those wire-to-wire latencies at the network switch level. And my average or median latency was around 40 microseconds. from message hitting my server to order leaving my server. Um and you know at tolerances of 40 microseconds with standard deviations of you know double digit microseconds, you have to look for every possible source. of delay. And you know, so you just look the at up and down the entire stack. You look at your networking equipment.

um from you know on your switches and your routers you look at the kind of lines that you're using whether you're all fiber or, you know, you have to sometimes you have to hunt down fiber to copper switches because those kill your n your latency.

You have to look at the network card that you're using. You want to use kernel bypass technology so you're not dealing with your operating system overhead. Um and that's you know then you have to tune all of your network card settings, your network driver settings, and all depends on the particular network. chip card you know, the chip in the card and the driver that you're using.

You want to tune your OS and your buffer sizes. And that's you know, that's before anything has even hit your own software, which you have to just profile and tune uh rigorously. And one thing I was able to do with that trading strategy was If I got an event that I wanted to react to and I tried to react to it and I missed, um I could measure the you know, the response time of that exchange on average to the times that I would get that I would get my fills.

And then I could measure the the the time from the market the public market data feeds of whoever beat me to that fill. To hitting that order, and I could profile my competitors and understand exactly how fast they were, and I would know exactly how many microseconds I had to take off. in order to to start beating them again. And it was you know, it was something that y you had to pay attention to. It would come up every week or two. It was this you know, sort of this race with each other.

Um and you know, I mean it was it was very difficult and and it was some really interesting and challenging engineering um problems, but you know it I think in the end you started to wonder what Why am I spending all this time trying to shave microseconds? What what exactly is that doing for the market? Mm. Now I've got a couple of questions based off of, you know, what you said there. So

The probably the first one is forty microseconds. Are you able to put that in perspective for us? Like, can you compare that to anything at all? Uh well it takes you three hundred milliseconds to blink your eye. Um and that's so that'd be three hundred thousand microseconds. it takes a nerve impulse, I think about on average eighty milliseconds to get to your uh you know, to your brain. So that'll be eighty thousand microseconds. So I you know, I

I'm not sure. I all my all my scales are in millisecond, you know, that we can relate to. So m yeah, microseconds, you know, you're talking millionth. of a second. So 45 millionths of a second. It's it's a it's a time scale that's hard to get your head around. And that was You know, that was in like twenty ten and twenty eleven. So they're you know, these guys now they're dealing in single digit microseconds, they're optimizing nanoseconds.

Um, things were headed that way when I left, but you know, that's you that's certainly where they're at now. Yeah, yeah. That's incredible. I mean, it's so hard to get your head around just how Hãy subscribe cho kênh Ghiền Mì Gõ

amount of timers, you know what I mean? Yeah, there's very little you can relate to. Yeah, yeah. So you said that you were like analysing and monitoring your competitive HFT firms to see you know, how many microseconds their orders were sort of hitting the servers and bouncing back out. And if other firms were just a couple microseconds, like a couple millionths of a second faster. Yeah.

Did that mean that you would lose if you weren't faster than them? I mean, I know that's that's probably a bit of a a newbie question, but I mean, was it really like that? For the the things that I was doing, it was pretty black or white. It was you know, an event would hit, you'd want to go clear out a bunch of price levels and you know, either you got it or they got it because the quantitative

side of what what we were doing was not, you know, dramatically advanced, I would say. I mean, you know, it was probably hard math for most people, but it's not like We always used to say it ain't rocket surgery, you know what we were doing. Um so it was very much about speed and and many um HFT strategies. Uh are speed dependent, and that's why you have this competition over lasers and microwaves. Now, back then, you know, it was

the spread networks line, the the Chicago to New York line um was a was a major change, a major shift. Um and if you weren't on it you know, that you might as well go do something else. Um, and that was a difference of like two and a half milliseconds or something like that. So You know, there are other strategies that are more quantitatively driven that are not as speed dependent.

You know, there are, you know, there's a a pretty substantial class of strategies that are speed dependent. And and I I should I should clarify often when I when I mean speed dependent, I mean yeah, it's usually not about who's second fastest. It's usually about who's the fastest. Now, that's for aggressive trading strategies. For more passive market making strategies, you know, it's okay to be in the group of the fastest.

But you know, as a, it's all about sort of Q position for the passive market making strategies. So, you know, as you are slower, it means your Q position is further and further in the back. Right. You're you're further and further towards the end of the line. And the closer to the end of the line you get

uh the more toxic the order flow that you're interacting with is, which you know, the more informed. It means that generally you're on the wrong side of the trade. So when you're at the front of the line, you have this backstop of a bunch of people behind you where if you get hit,

And you realize you don't want to be in this position, you can scratch out at whoever, you know, with whoever's behind you. But if you don't have that backstop, then generally, you know, if you get hit and you don't want to hold that position or it's it price is moving against you, you know, you're screwed, you're gonna lose money. And you and you talked about uh using lasers and microwaves uh in your response there. Uh how how are su how are such things being used?

So pretty much for market data transmission, although some people also send orders over those lines as well. Um but if you you know if you you can imagine that point to point data transmission is just it it's gonna be faster than a line that runs underground um that probably doesn't go in a straight line.

And also data traveling through the air is going to be faster than data bouncing around a fiber optic line, bouncing off the sides of the wire. So, you know, there's a pretty substantial latency difference. even if the the distance itself is the same, you know, even if somehow you got a fiber line that was point to point, traveling wirelessly is still going to be faster by a little bit.

Mm-hmm. A and these data transmitters like, you know, lasers or microwave towers or whatever you want to call them. Are these individually owned by said HFT firms? Uh yeah, you have some that are owned by individual HFT firms. There's a guy, um Who tracks a lot of this stuff? I think he's uh Belgium a Belgian anthropologist in uh Alexandre Le Manier and he does really interesting reports on

um wireless data transmission and whether individual firms own the towers uh or vendors own a bunch of the towers too. So um like McKay Brothers and I forget the the other the other firm. But you know, you have vendors that own these towers um and sell

the wireless spectrum, you know, bandwidth on the on the the towers and then you have some HFT firms that have just gone ahead and built their own. Okay. And and that's the same for some of the cables that are underground as well, is is that right? Yeah, that's true. Although most of the time it's you know, firms don't lay their own cables. Most of the time it's just it's vendors that own the cables or telecommunication companies that own the circuits. Okay, okay.

So, you know, anyone who's maybe seen you in the Wall Street Code or heard you speak in other interviews has probably heard your story about uh trading during the flash crash. But for anyone who might not have, you know, heard your take on on the the situation. Uh, would you mind sharing that story with those listening to this podcast?

Yeah, sure, I'd love to. You know, it was a like I I've said before, it was a pretty, you know, impactful event for me, really changed how I saw The impact of of high speed trading on the market and and just generally um, you know, the the complexities of market structure and where it was leading us to. Um, you know, it was It was just kind of a normal day at the time. You know, in twenty ten there was there was a lot more volatility. I mean now we're starting to see volatility, but

Um you know, for the last couple of years we've been in a pretty low vol environment. Um but in you know twenty ten that wasn't the case. I mean you can I got my start um in in actual trading in 2009. So it was a very, you know, vo very volatile couple of years. And so, you know, the markets they were down A couple percent, uh rioting in Greece, that kind of thing. It it it wasn't anything exceptional.

You know, our day was just kind of proceeding as normal and we were trading and I was working on new trading strategies at the time. Um And, you know, I don't I don't remember any more the exact times. I I'd have to look it up again. But you know, it was like around two twenty or so I look up at at C N B C and the Dow was down Um you know, whatever it was. And I I look up a couple minutes later, dropped another hundred points. And I said, Oh man, that's a that's a pretty quick drop.

I went back to what I was doing. I you know took a look at our training strategies, everything was going fine. Looked back up a minute later, was down another hundred points. My boss was sitting right next to me, he's a friend of mine, and I you know, I said, Hey, why don't you look up, look what's going on. And uh and he looked up and he said, Yeah, whatever you know, and went back to what he was doing.

I look up a minute later, it's down another 100 points, and I said, Hey, something's going on here. I think he looks back up, he looks at it, you know, we start looking at all of our training strategies. And we weren't um during the day, you know, we had again, like I said, uh some trading strategies, a lot of our stuff. uh at the time hadn't been really fully built out yet. We had we had moved

uh from one firm to another relatively recently. So it wasn't hard for us on the equities desk to shut down what we were doing. Um but, you know Well we looked over, as I said before, the the futures guys were right next to us and they were starting to scramble. You know, they were starting to to to panic and and to you know no one had any idea what was going on. And so the um you know the market dropped again and obviously kept dropping at at one point

the CEO of the firm came running over to the equity index guys and he was just, you know, shut it off, shut it all off. He started screaming. We ran over. We started helping them like close programs down. Everyone's just Shut it you know, every everything's just control C, control C everywhere you can go. Kill the programs, hit the stop buttons, you know, bunch of things have your big red button to just stop trading.

Um and then we're we all kind of at that point, no one knew what was going on. We all kind of gathered around one monitor, which just showed the e-mini order book. And as you're watching the order book. you know, the m of one of the most liquid instruments in the world. you just start you start seeing like price levels disappear and quantities just drop to nothing and and they you know the bids and the offers they start spreading out where it usually it's a very thick book.

And then they start, you know, moving away from each other. And suddenly the, you know, the book's gone. We're looking at it. What the hell is going on? There's nothing there, you know, and and we didn't know was it a data feed problem, what you know, was was something crazy going on. We look up at the TV, the market was down, you know, like crazy at that point. And and sure, you it's easy to look back and say, Oh, but then everything bounced back and you know, but that at that moment.

You had never thought this could happen. You had you had no idea what was gonna what was going to happen now. You didn't know what to do. You didn't know if you could trust your data or not. All the feeds were screwed up. Yeah, I I think I remember the SIP at that point was like minutes behind anything. And and so you just kinda sat there and and yeah, slowly

trader you know traders came back into the market, things started to move, we started turning things back on. But you know, there was that moment that I'll just I'd never forget where You know, you don't know if the market's coming back. You don't know what just happened to it. And um, you know, I I just I think it was that.

It was that thing, you know, that that just made me start to really question what was happening and and really as a firm it set us in a in a strange direction too. Um you know, we I remember the CEO coming out a a quarter later and he said, Oh, we're gonna get more involved in the public debate and make sure people understand what we're doing and see if we can help and

And so, you know, it was a real turning point I think for the obviously for the industry and for regulators and and yeah, it was a it was a dramatic uh way to experience it on on the floor of uh one of the largest e mini traders. Are you ready to get serious about trading? Then join Tasty Trade, Investopedia's best platform for options trading in 2026.

Stocks, options, futures, and more. Tasty Trade has everything you trade all in one platform. Get low commissions, including zero commission on stocks. so you can keep more of what you earn. Tasty Trade is packed with advanced charting tools, backtesting, a pre-built strategy selector, risk analysis tools, and more features to help you trade smarter. See equities and derivatives with high trading volumes, dividends, upcoming earnings reports, and more with their pre built watches.

Or create a custom watch list to keep an eye on the companies and sectors that matter to you. Manage your positions with speed and precision using Active Trader Mode one click trading and smart order tracking. Plus Tastytrad's Stellar Trade Desk team offers live support during trading hours if you need it. Visit Tasty Trade dot com slash chat for more info. KC Trade Inc. is a registered broker dealer and member of Finra, NFA and SIPC.

Yeah. Have you ever watched a stock explode and thought, if only I had the capital? Or sat on the sidelines because your account balance felt too small to matter. Good news. With Trade the Pool's limited risk platform, you don't need millions or even thousands to start trading the US stock market. Bypass the PDT and tap into over twelve thousand US licit equities. From penny stocks to big caps, ETFs, even the newest IPOs, and short anything you like, with zero locate or hard to borrow fees.

Start your evaluation, get funded with up to$200,000 in buying power so you can go big without risking your own savings. And now you can also have unlimited time to reach the profit target. It's a game changer. Not ready to trade yet? Trade the pool offers a free demo and educational resources. Practice on live data, master the platform, and build confidence risk-free before you even pay a cent. Click the link in the show notes to start trading with Trade the Pools Capital.

You know, there's there's still a lot of debate and a lot of confusion about what actually really happened that day and what sort of caused it. I mean

Do you have your own version of of what you think uh may have led to that event? I do, yeah. I you know, I'm a I'm a student of complex systems. It's something that uh I got really I was uh more of was just a passing hobby at the time of the flash crash, but you know, over time I've gotten much more into and and and spent a lot more time reading books about and and you know I'm I'm a strong believer that

that the market is is a complex system, has been for a long time. It's a different kind of complex system now than it used to be when it was human driven. Um and, you know, to me, there was no cause of the flash crash there.

It's that simple. There you in a nonlinear complex adaptive system, you don't have cause and effect. You know, you can you can be the regulators and point to the Waddell and Reed order and say, Well, that was what caused it but you know, you've had that those sized orders before and since.

despite being rare and they didn't cause flash crashes. So, you know, it's it's a very unique uh confluence of events that requ that you require to have something so dramatic happen and and so you know my preferred term that I've read in the academic literature um is this, you know, this this feedback loop or this illiquidity contagion. Um and so, you know, the system entered a positive feedback loop. and all these agents reacted in the same way, much like we did.

on our trading floor, we pulled everything. So did everyone else, you know, ever all the all the liquidity providers. Uh because the you know you didn't trust your data, you didn't know what was going on, so you stopped trading, and that's what happened. Um and it's you know uh a million individual reac individual actions that led to something like that. Um and you know, that's the kind of thing we see on a regular basis

on a single stock level now. And, you know, I I believe that a big part of that is that we need um a greater diversity of market making. We need market making on different timescales. Um and you know, there there are ways to encourage that from a regulatory perspective that in the at least in the US. w we don't go down that path. Um but in other countries, uh like Australia and Canada, uh you guys have and you know, I think that makes for stronger markets.

Yeah, no, that's that's really interesting. And I don't know if this is almost, you know, a little bit beyond the scope of the podcast, but would you be able to sort of explain what you mean by there there's no sort of cause and effect uh in in complex systems? Yeah, so it you know, it just means that In complex systems, you have emergent phenomena. You have things that that sort of just self-organize or just come out of.

c you know, i chaos is is a term for it. It's kind of a very specific term. It doesn't, you know, it it it doesn't mean, you know, sort of the chaos as we think of it. It's it's sort of a mathematical term. But you know, you have this this transition to turbulence. that happens in complex systems, which is, you know, if you've ever seen pictures of fractals, many many fractals are actually the visualization of

what it looks like, what what the math looks like of a transition from a system uh to turbulence. And so when a system transitions into turbulence, it it doesn't do so because something happened in particular. It does so because it's an emergent property of a complex system. It it's

It's kind of complicated, obviously it's it's kind of complicated to explain it, but um you know it's it's i chaos theory and nonlinear dynamics have a lot to say about this kind of thing. And it what it means is If you think there was an event that tipped a system into turbulence

And you can say that that event will always tip that same system into turbulence, that's cause and effect, that's linear causality. Um but if you know, you have the same event happen and sometimes it will push the system over and sometimes it won't, then you can't really say there's a cause and effect there because there's too much going on To model it in a linear fashion. You know, it it it has to be this dramatic confluence of events that happens.

Um that pushes it you know, that that that pushes the system over the edge. I I think is the you know, not not being really an expert in this, I think that's the best way I can explain it. I'm not don't know if that helped or didn't. Yeah, no, that was really good. And I think you I think that similar to what you explained in in the Wall Street code, I I think it was you that said that

You know, if all the same events were to happen again in the same order, you might get a totally different result. Was was that? That's right. That's exactly it. And that and that's the most that's the most important part.

You know, it's like it's sensitive dependence on initial conditions. It's the butterfly effect, right? So it's not that You know, may maybe it is modelable and maybe it is not, but the thing is that at our level, you know, there there are such tiny differences between what pushes the system into turbulence and doesn't that we, you know, we might think it was the same set of things that happened, but, you know, we're not able to model that.

That impact of something so tiny that we wouldn't even be measuring it. And yet it blossoms into such a large impact. you know, on a complex system. Yeah. And that's it's hugely fascinating. So

Following the flash crash, you kinda mentioned uh uh right at the beginning you had like this epiphany one day that led you to pretty much quit working within an HFT firm. I mean, can you tell us a little bit about what was going through your mind at that point? Yeah, um it it was interesting. I mean it was so in May of twenty ten was the flash crash. Um and I

I I I definitely got concerned and like I said, you know, our firm the the CEO came out a quarter later and he's like, Hey, we're gonna be involved in the, you know, public discussion, blah, blah, blah. So, you know Let us know. Let me know if you have any ideas on market structure or anything, you know, what what you're thinking. And so it was funny, I I wrote this whole thing up and I was like, you know, I think

We should advocate for these kinds of changes in the market. And a couple of the changes would have hurt the firm, no doubt. And I said, but. As a firm, if we're long term focused, if we're really interested in that. Um you know, long term strong, resilient markets are what are in our interests rather than maybe short term profits. So, you know, we should we should look at this stuff. And I I sent him, I emailed it to him and I asked to sit down with him and talk to him about it. Um

And he he did, and he sat down with me and he practically laughed at me. I mean I not pr I mean he did, he laughed. He like he looked at it and he like laughed. He's like, No, this isn't what I meant. This isn't what I was talking about. I was like, Oh. Okay. You didn't you didn't really mean what would help the markets. You know, it was how do we sell ourselves or something like that. Um so

You know, that was another event that that kind of set me on a different a different trajectory. Um I I kept working there for another few months and then uh my wife got pregnant for the first time and I said, All right, I don't know. I d I don't know how I'm going to, you know

tell my kids this is what I was doing or, you know, and and this is th this is really not meant to be a judgment on people who are in the industry, continue to do it. I have very good friends there. This was this was a deeply personal thing. And You know, I said to my wife, I said, I I just don't think this is what I should be doing. I don't think it's a good use of my time, of my, you know, intellect.

Uh I feel like I could be doing better stuff. Uh I I I came up with this analogy. You know, the the work was very stressful. You know, you're writing software that is responsible for millions of dollars going through in in very short timescales, bugs would be incredibly stressful, cost a lot of money. Um and I said, I don't know that, you know, you have a certain budget of stress in your life, right? You you can only have so much stress before you're done. And

I said, I don't know that this is what I want to spend my stress I said this, I don't I don't know what this is what I want to strengt my spend my stress budget on in my life. I feel like I could be doing more uh responsible things or you know, something that it made the world a bit of a better place, not to sound too uh bleeding heart or anything. But um so it was really funny uh by m by May or June of the next year.

I left and so um actually uh moved into my mom's house in New Jersey from Chicago and so there I was uh 31 with a pregnant wife living at my mom's house and unemployed and it was pretty sweet.

So tell us a little bit about what you do today. I know you're um your founder or co founder of Core Group. I mean, what sort of projects are you involved with there? Yeah, at at Core we we are um a market structure consulting uh shop, you know, I I got a partner who has great experience um also in market structure. He's been kind of a leading voice for years, uh Chris Nagy. We've got a bunch of programmers that work for us. Um in

We do all sorts of different market structure consulting projects. Um, but we have recently, and it's funny this this just is coincidental, but just uh today we're launching our new website um at KOR Trading dot com and we're really sh you know narrowing our focus. to what we call best execution consulting. So a lot of our work has become working for institutional asset managers, uh hedge funds, pension plans, mutual funds.

um, you know, uh long only's, that kind of thing, uh and long shorts. But you know, anyone who's pushing large orders through the market uh and needs to make sure that they're using sort of industry leading best execution practices, policies, procedures, that they're doing robust quantitative analysis. Um yeah, we've gone from a situation in which there was very little data on trading available to one in which there's way too much. And so now you have lots of firms at these large institutions

that are collecting all this data, analyzing their trading behavior, and they're overwhelmed by it. They don't even know how to, you know, what to make of it. And so we try and help them understand what their quantitative analysis means, making sure they're doing the right kind, they're looking at things in the right way. Um we hope that we help people reduce their transaction costs and you know, working for when you work for pension plans or mutual fund companies and you know that those

those cost savings go directly to people's retirement, that kind of thing. It you know, that's again, that's the kind of thing that I like doing. Um we do market structure workshops where we'll do a half day or full day session uh and and we'll completely customize it to the firm and you know that help them understand the history of market structure. How did we get here?

you know, where are we? What what are the complexities of market structure? What does it mean for you for you as a firm or as a regulator? We've done that that workshop for several regulatory agencies in the US.

And so that's the kind of stuff we do in our for profit lives. We also founded a non for profit called Healthy Markets. Um and at healthy markets we've got institutions who are members So we represent again the you know, institutional asset managers and the spectrum of small hedge funds that are you know, a little over a billion dollars to large asset managers like Janice or SEI that are hundreds of billions of dollars. Um and

With those with those asset managers, we help them again, we help them understand and keep them on top of market structure. We comment on new regulations, mostly in the US, but sometimes in other countries. We try and help regulators see things from the investors' perspective. We advocate within the industry for better standards around transparency and disclosure. We do rigorous examinations of ATSs and dark pools.

Uh, and we're now in the midst of building uh something that we call the Healthy Markets Research Institute, which is gonna provide a data platform. for academics and other people that want to study market structure where they can get public and proprietary data in an unconflicted way from an you know an independent provider for free um with public you know and proprietary data so that People can do studies on market structure with proprietary data, and then other scientists.

can replicate those studies, you know, and uh see if they can get the same results or if they can refute or support the results of other studies.

You know, that's a little thing I like to call science. And for some reason we don't have that in market structure. We have regulations that are being uh written right now, that are not based on data, that are not based on detailed study of what's going on in markets or that are based on conflicted studies and that's a dynamic that we're looking to change.

Yeah, right. Very cool. So you you said something there about uh, you know, how we've gone from a situation of having very little financial data to now having too much, like too much that we don't know what to do with it. you wouldn't happen to have any sort of stats or idea on how fast the amount of data uh that's available is is growing, would you? Um yeah, actually it's something I'm trying to work through right now as I

As I spec out the research institute platform, we just got a grant, so we're actually building it right now. Um I can tell you that. I was just talking about this yesterday with someone where if you look at opera data. and and um SIP data. So If you take peak data rates, which is what you kind of have to plan for, and you try to build something that can store though that kind of data, you need to plan for around 22 terabytes a month. And that's just top-of-book equity is an option.

Now, options is a maj is the major component of that. So when you take options out, it's more like 1.7 terabytes a month. Um but you know, so that's the scale you're talking about. Now with the research platform Uh we're planning on doing equities to begin with, although we will do futures and options quickly thereafter, but we want to do full depth of book equities. And for that, for a few years of data, you're talking over a hundred terabytes.

That's incredible. Yeah. It's a lot of space. It's a lot of space and that's why we um we have to partner with people like Google or Amazon. um to try and figure out how to make sense of that kind of data. Right, right. Okay. Wow. So

You know, while you're there at Core Group, I I know it was uh sometime in the past couple of years, you actually testified to Congress on the dangers of high speed trading. Uh, what were some of the issues that you raised uh with them? Yeah, uh so I've I've testified before the the Senate banking committee twice. Um and Іно бот там таке с перспективу да I think.

I l I like to think is very pragmatic. Uh, you know, I don't think that high speed trading in and of itself is a bad thing. I think that I'm a technologist, I'm a programmer. I think that the march of technology is inevitable and it's a good thing. I do think that some serious issues are are leading to unnecessary complexity in our markets. And a lot of those come down to conflicts of interest.

Um so you know when you look at exchanges and exchanges are part regulator and part publicly traded for profit entity. That's an example, a pretty dramatic example of a major conflict of interest, right? So who does the what should the exchange be optimizing for? Should it be optimizing its revenue or should it be optimizing uh the quality of its market?

And these things are often at, you know, not uh the same thing. So, you know, I I often talk about that fundamental SRO conflict. I talk about the conflicts with brokers routing orders. You know, what what should the broker be optimizing sh for when The client pays a fixed commission and the broker gets charged per share, depending on where it routes the order to. Should the broker be optimizing for its client's execution quality or for its own cost?

And you know, if it's optimized for its own costs, then it's gonna route to high rebate venues where it it sits there passively and most likely doesn't get executed versus you know routing for execution quality where it's gonna route to an inverted venue and have to pay but get executed and get better execution quality. So, you know, these are these these fundamental conflicts We find them in all sorts of places, um and often

They can exacerbate some of the negative aspects of high-speed trading, you know, of which there are some. And there's nobody in the industry that will tell you that there are no negative aspects of it, but there are a few in the industry that will tell you there are no positive. aspects to it also. I think that, you know, most trading at this point is high speed. And the question is Um you know, is it

this this kind of activity that's been called predatory, uh, which is aggressive trading and and structural arbitrage of the market. And that's a lot of what

you know, like it or hate it, uh some of the issues in Flash Boys were very relevant, were very accurate. You know, having worked at IEX, And worked with those guys, I can tell you that uh personally, you know, it was very a very good depiction of who those guys are and what they were trying to do and what we were trying to do for you know, I worked there for quite a bit of time and um

You know, there were structural issues in the market that had to do with market data feed, latency differences, and high speed firms. exploiting those differences and brokers catering to that exploitation. Um you know, so these are the kind of things when I when I go before the Senate, you you try not to get

deep deep into the detail, but you try and point out those high level issues that there are conflicts of interest, that there are structural design deficiencies, and that for some reason the SEC is not taking action. And and so you try and encourage, you know, Senate Banking Committee who holds the purse strings for the SEC, um you know, and has oversight responsibility, you try and encourage them to push these kinds of priorities.

Right, right. So can you flesh out some of the y you you highlighted there some of the high speed firms that are using what what mo some people might call predatory tactics. Can you flesh out exactly what some of those predatory tactics might actually be? Yeah, so you know, again, the Flashboys example is is one that, you know, despite all of the attempts to refute it, really has not been Refuted. And that is that if you don't route your orders in a latency-aware way.

Then you will see substantial order book fading. And this still happens. And you will hear many groups cite many studies that show that it doesn't happen and every single one of those studies is in a foreign market that's not fragmented.

So it's kind of a ridiculous analogy. Um but in the US, you know, you have a situation in which the NISE is in Mawa and the Azdax in Carteret and Bats is in um Secaucus and If you have a market maker sitting, resting orders in all of those order books with lasers between data centers or microwaves between data centers, and you as an institutional broker send a large order. you know, in without considering how it's being sent between data centers.

That order is gonna clear out a price level at one exchange, and it's gonna miss most of the liquidity on the others because those high-speed firms are gonna fade the order book in front of them. You know, I think that's an issue. High-speed traders will tell you that's not. They will say, well, that's us adjusting our risk, you know, in response to supply and demand.

I think that that's a little uh of a stretch. You know, it's one thing if your orders get hit across all exchanges and then you adjust, sure. Of course, no one's gonna begrudge that, but the idea that you can benefit from structural arbitrage because of high speed lines. you know, I don't see the benefit. I don't see the improvement in market quality that results from that. I just see increased volatility and uncertainty and order book instability. So

You know, I I think that's an example of a structural arbitrage. Um and when you dig into the data. Sometimes you see that it's more than just canceling. And that's when people start to talk about predatory. So It you know, it's one thing if you've got orders resting across these different exchanges and you pull them in response to something happening.

It's another thing if when you see something happening, you don't just pull your orders, but you start aggressively taking out liquidity elsewhere in order to move the price. against the order in transit. Um and so that's where you you go from order book fade to something that a lot of people would consider predatory and

When you dig into the data, you see that happening. You don't just see cancellations, you see aggressive orders racing ahead of institutional orders. Now there are brokers that have figured this out and And for whom it's not an issue. And there are are others that that haven't, that use less sophisticated order routers. Um and you know, so that's one example. Another example is data feed arbitrage, where if you know a venue is pricing midpoint orders based on the SIP.

And you have direct feeds, you know, you're anywhere from a couple hundred microseconds to a couple milliseconds faster than that matching system. And so you know that during price level changes you can exploit that latency difference.

And that's another example of a predatory structural art that doesn't enhance market quality. So, you know, I think those are the things that we want to be concerned about and that we want to address. And even if it's not a big issue, so even if the one that I just described is midpoint arbitrage, the data latency arbitrage is not a big issue, which many would argue it's not, and I still have yet to see data on whether it is or isn't. But it's a confidence issue and markets are

you know, predicated on the confidence of participants and that's why IEX has been so successful. You know, they those guys are on a mission and you know, like'em or hate'em or think that they're just marketing, whatever it is, you know, their business model is different. and they are really there to serve investors and that's That kind of confidence and trust has become rare in the industry.

And that's unfortunate, but it it does play a big role. Yeah, so I mean, we actually oh we I had uh Dan Azen from IX on the podcast uh a few weeks back and uh he was he was really interesting to to speak with. So what was your involvement with IX? Like you said before, you you worked there for a little uh I don't know, what was it, a a year or a few months or something? What what were you working on there?

It was about um eight months so it was when I just started uh being a consultant and like I said, in twenty twelve when that um that that story first went out on the radio. Uh Brad was one of the first people that contacted me. So I went up and I met with them um in the spring of or

early summer of twenty twelve when they were just getting started. No one knew who they were, what they were doing. Um, and I loved it. I loved what they were doing. I I joined up with them right away and I had a great time there for about eight months. uh just working with them on early designs. Um I helped to design the technology systems that uh you know that were ultimately built, you know, or to some extent built.

you know, and uh some of the anti-gaming techniques. I'm on a couple of their patents. And uh I just I loved it. I loved what they were trying to do to find a private market solution, not not to depend on regulators. Um I'm still close with the guys there and and you know, I I have a l um immense amount of respect and admiration for what they're doing, what they've done. And, you know, I think they're doing it for the right reasons.

And that's that's what was so refreshing to me. And I and and it's kind of part of a broader point, you know, like I said, I left the industry and I was kinda disgusted with it and and having come back in and and put myself out there and, you know talk to a lot of people in the industry about what I'm doing and why, it's it's been a really amazing and inspirational thing to meet a lot of like minded people.

Um, that you find out that there are all sorts of people in this industry. It's not just the caricature that you hear in popular press. And there are a lot of people out there that. you know, have a philosophy around trying to improve markets and understanding that what we're doing is is sort of a fundamental part of the global economy and you know that there are there's more than just

you know, a profit motive at work and that there should be. Um and and that's been a great part of all this and and also, you know, a a a great part of the work that I did with those guys. Yeah, no doubt. See, I mean you must have been thrilled to hear that um they were approved as an exchange, what, just uh last month, wasn't it? Yeah, yeah, absolutely. I I think they worked hard for it. I think it's it's a an innovation that deserves its chance.

Um and I I frankly and and this was in our comment letters, you know, we we didn't We we're we're a non profit. We didn't really want to take a stance on the issue, but we found some of the criticism to be a bit absurd. You know, the idea that three hundred fifty micro seconds would

uh ruin the NMS system or ruin the order book and make liquidity unattainable when you had latencies higher than that accessing NISE liquidity or far higher than that accessing Chicago. You know, it it just Some of the arguments were were kind of off the wall and and I thought really dramatically illustrated the split in the industry beh between investors. And i it was my count that I think four or five trillion dollars of asset managers supported IEX.

Um versus uh you know, their competitors and and a lot of proprietary and high speed trading firms and and that was a a pretty dramatic illustration to me of of what's going on in markets today. Okay. Okay. And as we know, IX was before becoming a an approved exchange, was a dark port or an ATS.

I mean, what's your views on dark pools? Like how does um what effect do dark pools have on listed exchanges? Like is there any harm that can come from executing off exchange? Oh I I believe so, absolutely. Um and you know one of the principles of our nonprofit is that we believe in displayed liquidity, we believe in open competition for order flow. Um You know, resilient markets are ones in which exchanges are the strongest part of the market that does there's ample incentive to display orders

um which order flow is not being segmented and siphoned away. Um and that you know that and and in which market makers are profitable. That's a that's what we believe. And so and and which a lot of evidence bears out and a lot of data and academic study bears out. So, you know, I think that when you have

uh dark pools being run by bulge bracket brokers uh or major order routers, that that's a pretty substantial conflict of interest, and there have been a lot of enforcement cases that have shown that to be true. Um, you know, hundreds hundreds of millions of dollars at this point of enforcement. Um and that when you uh

There that that there is a place for non displayed liquidity, both on exchange and off. And these these are very important and critical features to have in a market. But when you have, you know, twenty ATS um of any, you know, substantial volume and massive fragmentation across ATSs and disparate uh business models and fee schedules.

and internalizers that are siphoning off all of the retail order flow so that never makes it to an exchange. You know, you have all of these impediments to profitable order flow from a market maker's perspective getting to the market maker on the exchange. And to me that's an issue. That's what leads to f more fragile market.

Uh so we believe that, you know, um retail orders should go to an exchange uh like they do in Canada and Australia. We believe that dark pools have a role, but Um, you know, that it should be a much smaller role than they currently serve. Um and that, you know, all of this massive fragmentation is a bad thing. That there that there is a nice happy medium somewhere between, you know, the futures market, which is monolithic, and the the equities market, which is completely fragmented.

Uh and that regulation is sometimes needed to kind of help nudge the system in the right direction.

Yeah, okay. Okay. So, I mean, we should probably start to to wind this down. I mean, is there anything you want to add before we before we wrap this up?

You know, like anything you'd like to see changed moving forward? I don't know, is there anything you'd like to mention or do you think we've pretty much covered um enough for the time being? Yeah, I I think we we had a a bunch of different things. You know, I I I think that, you know, from a from me, from a a programmer and more and more of a scientific perspective, I I believe in data and data analysis and you know, I want to see markets

and regulation being much more data driven as opposed to these arguments that are filled with opinion and bias right now. And that's sort of what we're seeking to do with our with our nonprofit. And you know, I like to also empower people with information and help.

them understand the complexities of the market, uh, whether through stuff like this or, you know, when I when I work with asset managers. And so, you know, I think the more information we get get into the hands of everyone and then into the right people, the better. So

Um, you know, I'm I'm always open for more opportunities to do that and you know, happy to come back on if if people are interested and you know, if there are are more questions or you know, if if people wanna you know send me questions directly I'm always I'm always happy to to correspond. Absolutely. Well, where is the best place for, you know, any listener to connect with you? We'll find out more about you and and core group. Where where's the best place to go?

Yeah, you know, if you just wanna hear what I think, I I tweet a lot, as I'm sure uh people know, so that's my handle's just D Lauer. um at Twitter or if you want to email me it's Dave at KOR Trading dot com or you can just go to our new website, K Ortrading dot com, and there's a contact form there also. Good stuff. Good stuff, man. Well, I'll put all those links uh in the show notes as well at chatwithraders.com so you can find everything there all in one place.

Dave, I just wanna say, man, thank you so much for doing this. It's been um it's been awesome. It's been really, really fun. So I appreciate you coming on, man. Yeah, thanks for having me. I enjoyed it. the end of this episode of Chat with Traders. But rest assured there are more episodes. We'll catch you next time.