What’s Trading Got To Do With It?

Notes from Trading Show Chicago 2016

During an afternoon networking break at last week’s Trading Show Chicago event, I bumped into the COO of a clearing firm that I know. Beyond exchanging pleasantries, he stated that “I don’t know what I’m doing here”. With topics ranging from open source collaboration to the dollar value of a microsecond, the trading world that he once knew was gone. If Trading Show Chicago taught us anything, it’s that this is not your father’s trading world anymore.

Getting to the (Open) Source

The Trading Show Chicago, hosted by Terrapinn, was held over two days at Chicago’s Navy Pier and the first keynote signalled a different approach to trading than what would have been seen just a few short years ago. It has been a truism that financial services has been all take and no give when it comes to open source standards and technology, but Dirk Eddelbuettel from Ketchum Trading illustrated how that is changing. The fact of the matter is that trading firms have largely broken free of their knee-jerk paranoia about all of their tech and now recognize that they only need to protect those activities that generate alpha and not an activity that is a commoditized function. In the case of Ketchum, the 80/20 rule applies with 80% in commoditized functions like scripting, domain language, and storage and 20% devoted to edge strategies analysis and in-house trading technologies.

Lest you think that Ketchum is an outlier, both Brian Peterson of DV Trading, Peter Nabicht from 12 Sided Technolgy, and Michael Beal of Data Capital Management all referenced open source sharing practices in their discussions1 and Eddelbuettel commented on Two Sigma’s contributions to open source as well. In another case, I spoke to a rep from one of the largest prop trading firms in the world and he told me that they have been very generous with financial support for open source groups, although their contributions remain anonymous. I suspect that taking funds from a prop trading firm might rub some people the wrong way.

Panels and Presentations in a Nutshell

Over the next day and a half, the panels and presentations boiled down to one simple fact: it’s all about the tech, ‘bout the tech, ‘bout the tech, in trading (apologies to Meghan Trainor). Here’s a brief recap on some of the highlights:

• Brian Peterson, DV Trading: in addition to his comments on open source, Peterson examined the disconnect between academic and industry in high performance computing. Take-away: the scrappy, real-world of trading can teach a thing or two to academia about improvisational flexibility in technology.
• Next Generation Infrastructure: this panel spent too much time on the hoary topic of “build v. buy” (hint: it’s “buy”, per open source comments above) with the biggest reveal being how pervasive cloud has become, particularly when it comes to analysis and testing. Another myth dispelled: trading loves the cloud.
• Stephane Tyc, McKay Brothers and Mike Persico, Anova Technologies: In separate presentations, these connectivity providers illustrated that the race to zero is still being run. Tyc described the specifics of connections between the CME in Aurora and NYSE in Mahwah (they’re within 32 microseconds of the speed of light2), how they were able to reduce the time between Basildon and Frankfurt3 by a few microseconds by drilling through a wall (although the telco doesn’t know about it yet), and how large U.S. prop firms are hoping to build “Eiffel Towers” on the coast of England to get the shortest path possible between London and Frankfurt. Persico, for his part, went over the methodology that is the basis for an upcoming white paper estimates that the total latency alpha between NYSE, BATS, and Nasdaq is $758 billion and that the value of a microsecond is something on the order of $7.724.
• Matthew Dixon of IIT / Quiota and Michael Beal of Data Capital Management: While the need for speed is still acute, the dive into data is progressing as well. Dixon described how deep neural networks are already making an impact in areas like speech recognition and self-driving cars as well as in financial markets. Meanwhile Beal described how his firm is akin to “Google and JP Morgan giving birth to a hedge fund”, one that is driven by traditional and nontraditional data and tools.
• Tech, tech tech: Programming with R (Jeffrey Ryan, Citadel), cognitive computing (Josh Cohen, IBM Watson), NVMe accelerated caching (Brad Winett, DataDirect Networks): need I say more?

More to Trading than Just Technology?

To be fair, the whole conference was not focused on technology. The first day had a “Quant World & Big Data” track that touched on more traditional topics like dynamic portfolio management (Attilio Meucci, KKR), the impact of rising interest rates on low-risk investment portfolios (David Jessup, UBS) and the relationship between index and individual stock volatilities (Brian Liang, Bloomberg). However, the discussion came back time and time again to technology. For better or worse, that’s the way the trading world revolves these days.

And last but not least, no financial conference these days can go without a discussion of blockchain and The Trading Show was no exception. In this case, however, I was the moderator of the panel so I don’t have notes on the comments made. For that, eFinancialCareers filed a story: you can read it here.

1Another example of open source use is Spot Trading, as detailed in a posting on the FinTEx Chicago website.
2A microsecond is one/one-millionth of a second or 0.000001 seconds. One microsecond is to one second as one second is to 11? days,
3Tyc also recommended the blog “Sniper in Mahwah” for deep-in-the-weeds analysis and discussion on connectivity issues. It makes for interesting reading.
4At least that’s what I have in my notes: it was the last presentation of the day and my head was spinning by that point.