Fast Money: The Potential of Speed (Part 1)

The New York Stock Exchange; it’s the iconic backdrop of all things financial and is a beacon of the complexity & supremacy of the US economy. But unknown to many, about 70% of stock trades occur off the NYSE trading floor and the majority of these occur through high frequency trading.

As the name suggests high frequency trading is fast as in extremely fast. Companies such as Goldman Sachs, Morgan Stanley, Merrill Lynch etc. that engage in this kind of transaction often buy or sell over a billion shares every day which to put in context, is over 11,500 stocks every single second. This outrageous volume is handled by computers governed by complex algorithms that take market factors into account to predict minute fluctuations. Despite the wealth of information processed, the end goal is simple; as always, buy low and sell high. Now high doesn’t have to be very high. In the example of the TradeWorks LLC, computers in charge of high frequency trading are instructed to make a profit as small as a penny or less 40 million times every day. It’s easy to see that the pennies really do add up as the relatively small corporation nets approximately $400,000 every day solely from this activity. Don’t forget that these numbers are based on 40 million stocks a day rather than the billion stocks daily of larger companies, mentioned above.

But in many cases, merely knowing the algorithms behind successful calculation is not close to enough. As TradeWorks LLC CEO puts it, eventual profit lies in the ability “to capture opportunities that last for only a fraction of a second” often less that the blink of an eye. This doesn’t require any “special” Wall Street knowledge as many falsely suspect but does entail knowing publicly available information long before it reaches the public. For this, companies adopted the most advanced computers and communication systems available. Yet as this sector becomes increasingly competitive, collocation has become popular in which traders offer tens of thousands of dollars every month to place their computers right next to exchange servers.