As the trading landscape has undergone significant changes over the last several years, financial institutions have developed an insatiable need for low latency. Infrastructures that once had to handle a maximum of 50 trades per second, now must process thousands – an increase of more than 1,900%. However, aiming to gain the lowest latency possible should not necessarily be your firm’s goal. To succeed in the current environment, you need to have a good understanding of latency within your own infrastructure first and not simply engage in a blind pursuit of ultra-low latency.
Depending on your company’s trading goals, obviously, having the shortest line from point A to point B can be a major advantage. However, it opens up new, more complex strategies that are not ideal in a high-latency environment and can also slow you down if used incorrectly.
Be an Educated Latency Consumer
In order to find the right latency strategy for your firm, consider the below three points:
1. Be educated – identify your ideas and strategies, outline the execution and the final goal.
2. Don’t be held hostage to the hype – ultra-low latency may be the topic du jour, however, it might not be right for you.
3. Learn to manage speeds – in order to reach optimal solution you need to be able to adapt and be open to exploring all options.
Firstly, you need to map out the latency that is built into every platform and every intermediary system that your orders go through. Once you have a better picture of the route your order takes from entry through execution, you can start to manage your direct market access (DMA), which gives you the ability to mix manual and electronic trading strategies based on your ultimate investment goals.
Remember, no matter what you do, the drive to cut latency is here to stay. With this in mind, you want to make sure that you do not become a hostage to latency and the illusion of unachievably high speeds. Most firms don’t need low latency, they are just position takers that want to say they have it for the sake of having it.
There are some fund managers that pay a lot of money for an ultra-low latency infrastructure, but in reality all they need is relatively good latency. They can get that through one of the intermediaries that come very well equipped and “free” – that is, paid for through their commissions.
There are a lot of misconceptions about low and ultra-low latency – people think that they can beat the speed of light, which is impossible. When you’re situated in the US and routing through Tokyo and Europe, your order takes longer to deliver and you have to factor that into your plan. That means, potentially, putting an office in these regions. But either way, you need to know if you’re trading long-haul, over a significant stretch of network, when you will lose execution speed. Either de-concentrate your trading or integrate what you know about latency into your trading strategy.
Finally, if you have all these elements, it comes down to managing latency properly. This means understanding what you are facing and knowing how to leverage it to your advantage. Don’t try to build from scratch if you can already find a high quality algorithm integrated with latency at the core. Know what everyone is offering and adapt that to your own needs for the ultimate results. When you’re deciding on the latency of your infrastructure, tread carefully. The right solution and answer varies for everyone, there is no one-speed-fits-all solution.
