The question comes frequently from venture capitalists interested in an investment, sometimes from trade analysts or members of the tech media, but most often it comes from within the 25,000-member Marketecetra community itself.

Usually the person asking the question is a new member of the community that has set out to build a new trading platform for their firm. They have selected Marketcetera as the core of this new platform in an effort to accelerate their time to market and are often looking to collaborate with other community members and take advantage of existing best practices.

As part of this effort to better connect community members, we have surveyed new registrants to understand their composition and priorities. If you contact me directly I’d be pleased to share the complete results, but there are a few points that I would like to highlight.

First of all, 63% of our community members work for early stage firms looking to establish a new trading platform. This is consistent with our ad hoc conversations with community members, who we have found are often veteran traders and developers working for an early stage firm, often while simultaneously raising money and juggling a host of other responsibilities.

These firms are universally uncomfortable with the closed source, proprietary model that requires them to trust their business to a trading platform without the ability to review the quality of the underlying source code. This is especially true of high frequency traders leery of poorly written, patchwork code or business practices that require them to expose their strategies to proprietary vendors for coding into their platforms. The open source model, alternatively, provides tremendous flexibility and control. Clients have the ability to download and test our source code before committing their strategies (and business) to the Marketcetera platform, they decide which components of the platform they want to use and they have the ability to do their own customization or integration work, involve a third party development company or use the Marketcetera consulting team.

Second, budgets are tight and community members are looking to leverage open source (both the expertise of the community and the source code) to increase the efficiency of their operations. As early stage firms they are focused on keeping operating costs in check while they establish a track record of trading success for their investors and they are not willing to pay expensive license fees to proprietary vendors.

Fully 82% of our community members are deciding between developing a platform based on Marketcetera or building everything from scratch. Although they have extensive technical skills, they are skeptical of building an entire trading platform from scratch and even more concerned about supporting and maintaining a homegrown effort over time. The real issue is core competency and they would prefer to invest their development skills (and time) into revenue generating activities like coding new trading strategies instead of building and maintaining “plumbing” like market data adapters and FIX connectors.

Finally, 50% of the members we surveyed indicated they are willing to contribute back to the Marketcetera community. Historically the nature of community contributions vary, sometimes actual code snippets or features, often q/a work or answering forum questions, ALWAYS they are willing to share thoughts and opinions on what we should tackle next. The point is that the community represents a resource that developers and traders can leverage as they decide on a new trading platform or to accelerate their efforts once they decide to build on the Marketcetera open source project.

I hope this context on our community is useful. If you find yourself grappling with similar issues or you would like to become a community member, please check us out at www.marketcetera.org.

Download. Run. Trade.

In the midst of this full-throttle pursuit, it is worth considering some of the fundamental assumptions of great software design. One of the most basic (and often over-looked) questions is whether to pursue a proprietary system or opt for an open source approach.

History provides some useful context. In October 1991, a group of volunteers released Linux, a free operating system. The most recent release of Ubuntu Linux, one of the most popular distributions, is comprised of 100 million lines of code. As a point of comparison, Microsoft’s much postponed Windows Vista operating system was approximately 50 million lines of code. The software industry is strongly influenced by an engineering culture that values tight centralized planning, control and monitoring of the development process. How could a volunteer effort deliver a system twice the size of Microsoft’s offering?

The answer, I would argue, results from the “sunlight” benefits of the open source model. Development teams are organized not by management teams, but rather by the developers themselves, with individuals free to gravitate to tasks of personal interest and passion. This is coupled with the tremendously valuable process of peer review. Unlike proprietary systems, the open source projects freely publish their source code for other developers (and customers) to review. As Lerner and Tirole have documented, the open source model works especially well in communities—for example, software developers—where there are strong “reputation effects.” Instead of hiding the software’s source code as a company’s proprietary secret, open source programmers publish their code on the Internet for others to read, learn from, experiment with, modify, improve, and reuse.

The effect of this “openness” cannot be overstated. Today’s financial services industry is awash in a tangled alphabet soup of proprietary technologies, few of which are designed to integrate smoothly. In fact, many of these systems come from competing software companies with little or no interest in ensuring that their products work with those from their competitors (and often quite the opposite).

The complexity of today’s systems, the number of products we integrate, the speed at which we trade, and the secrecy that cloaks many of our operations threaten to create inherently unstable systems. Understanding the infamous “Flash Crash,” for example, consumed months of work by both regulators and industry participants and it has still become a touch point for people worried about the stability of today’s trading infrastructure. Outside our industry, both Airbus and Boeing provide object lessons on the limitations of the central planning model. Open source lacks the marketing muscle of richer proprietary cousins, but our model presents a genuine alternative to current industry practice, one where the balance of power resides with trading firms that gain the leverage to interrogate the source code of their most mission critical applications before they are deployed.

High frequency trading has gotten a lot of media attention over the past few days, much of which most of which has been negative.  This is due to a couple of factors:

1. The down market has created disgruntled (and vocal) concerned citizens wary of the finance industry.  That resentment and distrust has surfaced in the mainstream media in which bailout money, AIG, Lehman Brothers and the finance industry are lumped together and demonized as an amorphous, sinister entity out to trample the rights of the common person.  High frequency, in the eyes of the layman, is representative of this.
2. People don’t really understand high frequency trading.  In many instances, it’s wrongfully lumped together with practices like flash orders or short selling.  The real danger with this lack of understanding is that it leads to uninformed regulation. This perpetuates the lopsidedness of the market and results in regulations created to control a few being extended to many. One size cannot fit all in a regulatory landscape.

What is the Next Hurdle?

The market crash has created a political environment where heavily lobbied decisions that may not in the best interest of market participants are being forced on the industry.  Under political pressure and widespread scrutiny from the media, the SEC doesn’t have the luxury of time to do extensive due diligence before legislating.

During the the crash the SEC tried to prevent stocks from falling with regulations like banning short selling, but the spreads only got larger and because of the hysteria, there is now political pressure to bring back some version of the rule which was studied for years before finally being repealed.  The data and statistics show that it will probably have a negative effect but there is a hue and cry to do something, so something must be done. For those who argue that current market structure does not work need a lesson in history, to start of with even with all the turmoil that was happening in 08 stocks still have tight spreads and liquidity was available across the broad spectrum from small cap to large cap. What more of a test can you ask for than what happened in 08 ?  Whereas back on Black Monday in 87 and in the early 90s, for Nasdaq listed stocks the market would completely dry up the bid/ask spreads become too wide to be usable in really bad market conditions.

The current climate, lack of knowledge about what caused the market crash and the need to do something combined with a lack of understanding of HFT has caused it to become a the target of blame for anything wrong.

What Does the Future Hold?

Matching engines, a major part of the trade execution process, are currently limited by the communications between servers they run on. Technology-forward liquidity venues like NYSE are working on things like broad in-memory trading.  In as little as 3-4 years, everything will sit in one big memory box, removing the physical drag of communication between computers.  Latency’s only limitation will become that pesky speed of light and it will make today’s trading pace look like a horse and cart race.

The cost of entry and ability for smaller players to be able to participate will continue to increase as more competing vendors force prices down, for example the executing broker Lime Brokerage is known for providing very low latency execution.

Technologies like open source are also enjoying extensive adoption at investment firms of all sizes. Access to source code means infinite flexibility without being beholden to proprietary vendor release schedules and cost. Financial firms that have no internal software development can take advantage of the existing functionality in an open source platform, and innovate where they need to. Some of the world’s largest banks are using open source at every point in the investment cycle including Linux platforms, CEP engines like Esper, front-end trading platforms like Marketcetera and feed handlers like the Collaborative Software Initiative.

What Does it all Mean?

Several years ago, the barrier to entry into the world of high frequency trading was patently higher than it is today.  There were numerous startup costs, including expensive hardware, implementation, service and maintenance.  Only the bulge bracket sell side and buy side firms could play in this game. Today, if you are a small buy-side shop with two traders and a developer you could be up and trading automated strategies productively in three to four months.

There is a also a cultural change starting whereby technical people are collaborating virtually and sharing information about the standardized infrastructure… a great example is QuickFIX.  Trading firms are able to spend more on intellectual capital – the strategies and the algorithms that will gain them competitive advantage – instead of saddling themselves with large capital investments in expensive and restrictive infrastructure. The level playing field created by this perfect storm of technology and market pressure will bring with it thousands of new entrants into the high frequency trading game.

Ask a dozen traders and you’ll get a dozen different answers.  Some say it involves program trading, stat arb and quantitative modeling coupled with extremely low latency and high frequency trade execution. Others say it is where alpha generation is purely based on speed.

Quants define it as “exceptional speed + algorithms or automatic strategies.” But, who gets to determine whether your speed is “exceptional” or if you are using enough algorithms to make the cut?

There are even nuances to the technical definition. Does trade execution have to occur for it be considered high frequency trading? What about bad liquidity – if you put an order out to try and influence other participants, but you pull it back and don’t execute it, does that still count? What about sponsored access? Are you a high frequency trader if you are simply the conduit, not the execution venue?

The truth is that as an industry, we may never standardize on a definition.

However, a few things do seem to be constant. You need a fully automated system to be considered a high frequency trader and most people agree that it has more to do with the percentage of total volume traded using high frequency strategies than the total volume traded overall. This is interesting because it puts companies like Lime Brokerage in the same category as Goldman Sachs despite the fact that they are on opposite sides of the size spectrum.

How Did We Get Here?

To really understand high frequency trading, we must understand where the industry has come from…  Advances in computing power have boosted the number of firms who are able to take advantage of pricing inefficiencies in different markets. Changes in liquidity providers – particularly new entrants like Knight Capital, Liquidnet and BATS – have opened the floodgates for many high frequency firms.

Before decimalization, you had to be registered with an Exchange to be able to make a market as an on the floor market-maker (NYSE specialist) or an “upstairs” market maker (eg. Nasdaq market-maker).  As a market-maker you were subject to regulation and capital requirements in return for a right to capture the bid-ask spread. Before decimalization only registered market makers were allowed on both sides of the trade (eg. Bid and ask) at the same time. Now anyone can operate from both sides of the market and act like a market maker. This freedom, combined with the democratization of access to new technologies, has created a host of opportunities.  In 2001, one to two seconds was fast, but the continual increase in computer power combined with decreased cost has led to an arms race to see who can have the fastest and lowest latency infrastructure. In a market structure that determines the priority of orders by price and time, being first in line means getting the trade done.  This ever-increasing speed means that turn around times are now measured in single milliseconds or sub-millisecond for the largest players with co-located servers at the exchanges.

People have also been leaving larger firms to start small shops. And, by capitalizing on advancements in technology, they have been able to achieve the same results. This is currently resulting in a very large increase of newer, smaller firms competing with similar strategies and ideas. As is usual with the markets, even the small inefficiencies found today will soon be gone and spreads will continue to narrow until they are all at 0.01c.  This is an indication of a competitive market and good for retail investors as they will have lower transaction costs due to the decreased bid/ask spread.

First, hedge fund returns are up. And not just up from the carnage of last year, but with 2.5% returns in September, it’s looking like hedge funds could have their best year since 2003.  According to the Eurekahedge Fund Index — and bolstered by a number of other hedge fund research firms — the industry grew by $26.3 billion last month, bringing total assets managed to $1.42 trillion.

Such news, coming just months after record losses, illustrates that hedge funds are coming back aggressively and when you dig into the numbers, you find that funds with medium or high-frequency strategies had even higher returns.

Now to that other data point. Hedge Fund Research is reporting that hedge fund creation is up but the size of those hedge funds is down. And while it’s true that the new funds will have to compete with the larger established players for investment funds, we see this as a positive trend for the industry.

The reason why is simple. Bigger isn’t always better. Just look at the economic data that show that small and medium businesses provide the bulk of our jobs and most of the muscle behind a recovery.

In the case of the hedge fund industry, we’re seeing a resurgent wave of creative capitalism rising at the hands of small groups of traders who are making smart use of technology to leverage their trading strategies and secure returns. Many of them came from the big funds that posted last year’s dismal returns. They learned from the experience and are taking a new approach this time around with funds that may be smaller but are also more nimble.

We’d like to think that it’s this generation of hedge funds that Marketcetera’s trading platform is made for…

Unfortunately HFT is a term like “salad” where there are many definitions, and no one covers the entire set.  We define HFT as high-speed, high-volume, low-latency trading enabled by technology. It’s what the Marketcetera platform enables and it’s all about responding quickly to inefficiencies in the market..

It’s also unfortunate that the media have helped conflate HFT with a couple of other trends that have given it a bad name. The practice of flash orders — for example, giving order floor access to a limited number of market participants, which I would argue isn’t good for the market structure — for example, is being twisted up with HFT in some news stories, coloring it an unfair advantage.  It’s true the same technology can enable HFT and trading flash orders, but the two are not equivalent.  There are many high-frequency strategies that operate outside the limited number of venues providing flash orders.

A second issue conflated with HFT in recent discussions is the apparently large profits generated by banks that a) have taken TARP funds, and b) have produced significant returns based on HFT.  HFT has been identified as a government-funded unfair advantage that the big banks have over the individual investor.

I don’t buy it.

Whenever inefficiencies can be removed, it’s constructive to the market. That to me is a fundamental principle of capital markets. And while it’s true that the larger banks are more likely to have the technology budgets to develop the infrastructure to support HFT, the playing field is leveling quickly.

Bear with me for a marketing message: because Marketcetera’s trading platform is open source, it is cheap, fast and flexible. Our new offering in partnership with NYSE will provide a managed service version of the Marketcetera platform right there in the exchange’s data center, allowing for the same kind of low-latency trading that the big banks, who have paid to park their servers their too, are building strategies.

The point is, technology is evolving and Wall Street is changing. HFT is no longer going to be a tool for the elite few. The analogy we like to use is the racetrack where all drivers have the same car. Once the advantage of the machine is removed, the best driver wins.

At the SIFMA Technology Management Show this week, Marketcetera announced a partnership with NYSE Technologies to offer a SaaS trading platform over NYSE’s Secure Financial Transaction Infrastructure (SFTI) network. High frequency traders using the network to connect seamlessly with the world’s leading markets will be able to employ Marketcetera’s trading platform hosted in the cloud. This means high-performance, end-to-end trading infrastructure available at a fraction of the cost of proprietary IT systems that must be run on premise. Combining open source with on-demand services in the cloud will revolutionize the way that customers mold the Marketcetera platform to fit with their business models.

To help us express our vision for the future of open source trading and innovation in hosted cloud services, we’ve added a new video to the YouTube channel featuring interviews with Graham Miller, our co-founder and CEO, and I. Appreciate your thoughts!

Marketcetera Announces Partnership With NYSE Technologies

In the past few weeks, we’ve started with three basic videos: the first screencast describes developing simple strategies in Ruby on Photon, the integrated Strategy Studio. The second screencast shows how to deploy a number of real sample strategies. The third screencast is a demonstration of the Strategy Agent, which allows users to scale strategies to work on multiple machines and handle complex data flows between strategy modules. Future screencasts will include more complex topics and trading strategies. For example, you may see a video on pairs trading and a demo on backtesting.

The goal for the video series as a whole is to provide Marketcetera users with audio-visual explanations for all the features and functions of the Marketcetera platform. We’ve been listening to your feedback so far. Keep it coming! You’ll help us determine the future direction of the videos.

One of our goals at Marketcetera is to make the platform intuitive and easy to use. To that end, this screencast is meant to enable current and new users to learn the ins and outs of the Strategy Studio more quickly. As always, we appreciate your feedback on the product as well as our first screencast.

In reality, “automated” trading works very differently from this scenario. In fact, automated trading platforms give human buy-side traders more control over trading strategy and execution and allow them to adjust to the changing market in real time. To understand how this works in practice, let’s take a quick trip back to the time when the buy-side started adopting automated trading systems over a decade ago.

Before the age of automation, there was a distinct separation between the activities of the buy-side and the sell-side in the financial services world. The buy-side gave trading instructions to the sell-side and waited for the sell-side to execute these instructions in capital markets. This was a slow process. By the time the sell-side firms executed trades, the circumstances in the market may have changed to the point that the original instructions were no longer desirable from the perspective of the buy-side. As trading volumes exploded, this paradigm was no longer sustainable.

To cure the ills of a slow, disjointed trading process, the buy-side started to adopt technology. What if buy-side traders could not only develop sophisticated strategies for trading, but also automate the process of performing those strategies by themselves? With the help of technology, the buy-side could determine what to trade and when to trade it. With the help of technology, traders could have direct market access and turn quantitative models in their heads into real time executions. Since hedge funds deal with particularly complex strategies, they led the charge in automation. This put pressure on other asset managers to keep pace and a technology race ensued.

Today, many buy-side firms and traders work closely with automated trading technology to couple greater control with super fast execution and analysis. Qualitative traders use future-looking announcements from companies and quantitative traders use historical and statistical data to create models and mathematical algorithms for deciding what to trade and the opportune moment to trade. With automated trading systems, both types of traders can launch these models via the technology, tweak them as new data becomes available, and execute trades immediately. This means direct market access. No unnecessary, costly delays. The trader drives the system, but the system is powerful enough to act effectively upon the trader’s vision.

Are there any potential risks to this approach? Sure, markets are complex and the automated system may encounter events which the trader had not considered, but this risk can be mitigated by a flexible system that allows the trader to intervene easily.  At Marketcetera, we built our automated trading platform with this issue in mind. Average traders and large buy-side firms alike can benefit from an open source platform that is easily accessible and moldable to specific requirements in real time. The human factor is still critical in the trading world, and humans require freedom and control to maximize success in the market.