It has been more than a month since Michael Lewis released
Flashboys, an adrenaline shot in the arm of the automated
trading industry. But despite the picture that Michael Lewis
paints, trading – and automated trading in particular
– is not all about speed. In fact, in
today’s dynamic environment, the quest for zero
latency provides global trading firms with incrementally less
marginal returns. Simply stated: latency’s
low-hanging fruit has already been picked.
That isn’t to say that some firms
won’t find a competitive advantage to being
faster; rather, it is clear that for any given $1 investment,
it is better to dedicate resources towards finding smarter
trading strategies as opposed to faster ones.
Invest in new markets, asset classes
30 years ago, traders were limited by what they could trade
based on their physical location. If they were in the corn
pits, perhaps they could wander to an adjacent pit, but they
remained quite limited. Today, the same trader can trade any
number of products, virtually 24 hours a day, from the Asian
open to the US close.
In this environment, firms with adaptable software and
technologies benefit. Imagine a profitable strategy deployed in
a single product. Today, the most successful firms are the ones
able to take these working trades and quickly scale them out to
new asset classes or geographies. With arbitrage opportunities
often small or fleeting, yet increasingly expensive to
discover, the ability to replicate what works to many venues
ensures the highest possible return on research dollars.
Traders need software that can offer this on/off
functionality. Using the right software, traders can process
more data simultaneously, running strategies concurrently.
Scaling horizontally, from US short-term interest rates to
European STIRS, for example, is often as simple as checking or
unchecking a box.
Speed matters – in iteration
Markets offer ever-changing dynamics, and the quicker that a
modern, global firm can go from idea to production, the better.
Similarly, the faster that one iteration can be morphed into
another more profitable model, the better. Today, firms are
shortening development times and tweaking successful automated
strategies, instead of sending them back into simulation for
another full round of testing.
This is important as these traders and firms are able to
iterate quickly and adapt quickly. Focusing on a niche in the
market helps to make that possible. From there, traders are
able to move that strategy to new markets where it may also be
Diversify the trade environment
The proliferation of high-quality data analysis tools means
that many traders now use multiple applications to discover
trading ideas. Often, the best tool for sifting through data or
performing large matrix calculations is different from the best
execution platform or risk screen. Consequently, the trading
environment today has transformed into a collection of software
knit together by communication channels called APIs.
By using software designed to communicate in well-known ways
with other software, traders can easily choose the tools that
best fit the needs of their business. At the same time, the
'swappability’ inherent in such a modular system
means that as users encounter new markets and new products that
have different requirements, they can simply switch pieces of
their system without replacing their entire environment. As a
result, traders get better tools that are more adaptable and
more scalable to their needs.
If Mr Lewis were looking for a theme to today’s
global automated trading, this is what he would have found:
firms are committed to trading smarter, not simply faster.
While speed does matter, there are many things much more
important. Unfortunately, that realisation would not have
provided him with the villain that he was looking for.