Help Center Blog Open An Account

Trading automation with discretiony elements


#1

I want to ask about the opinions of traders here who use automation in trading but integrate an element of discretion. If so, when would you use such an approach. I still find it hard to let my system go “hands-free” because of risk management. When it comes to risk management which approach is correct when managing a portfolio.
G


#2

You have 2 questions in one, so one question is whether automation should work on it’s on, and the integration of risk management into trading. I personally think that system should run on its own without interfering, however there could be those that oppose to this opinion as well. Practical input of system provided by those who have actually committed real money would give you a good idea of what sounds good in theory and what is applicable in real time trading involving real capital.

I would like the address the issue of risk management and its application whether via dictionary of automation. You have 2 questions in one, so one question is whether automation should work on it’s on, and the integration of risk management into trading. I personally think that system should run on its own without interfering, however there could be those that oppose to this opinion as well. Practical input of system provided by those who have actually committed real money would give you a good idea of what sounds good in theory and what is applicable in real time trading involving real capital.

I would like the address the issue of risk management and its application whether via dictionary of automation.

While rules-based trading systems, and strict risk management is encouraged, it’s important to look at more than just the numbers when it comes to building, deploying, monitoring, and maintaining a strategy. Whether discretionary or automated, a trading system should always have contingency plans for emergencies.

Natural disasters, government announcements, and many more examples can cause short-term market fluctuations that may not be accounted for within the code of a strategy. Rigid monitoring and supervision should be the norm, even for completely automated strategies. Staying out of the market at a time where uncertainty is heightened can be crucial to capital preservation.

The number one focus for any successful trader is always to manage risk in an appropriate way. Consider that trading is a marathon rather than a sprint. When profitable traders rise above the rest, it is generally because their profits are achieved over series after series of trading sessions rather than gambling on a single trade. When risk is managed correctly, it can extend the life of a trading account.

Some trading systems have a short life span while market inefficiencies are corrected. Recognizing when to pull a strategy from live deployment is also something that requires much thought. Many traders become attached to a strategy and refuse to pull it from their trading rotation. Unfortunately, these cases often result in losses becoming greater than profits, and morale is brought to an all-time low.

This means that having contingency plans for short-term termination of a strategy should also be combined with a long-term plan for a strategy. Reviewing strategy performance versus expectation and exposure can provide a way to continually optimize and maintain the life of a system.


#3

A good system works off of probabilities. The problem with using discretion at times is you essentially change the probabilities per say of the entire system. You are introducing a variable into the system that can’t be measured in a way that can be relied on. Your discretion may work at times but not being able to measure its over all effect makes it difficult to validate your system.

If there is something that would for instance cause you to turn system off and on make sure it is something you can measure or test. So you can see how it effects your overall results over time. I think it is a bad idea to “think” hmmm I just don’t like the way the market looks today. Some of my best days have happened on days I thought I do not like the way the market looks or feels. For me to intervene with my system is kind of like putting a glitch in the Matrix.

Develop a good system then let it do its thing.