Live Test Update: We're Up $20K!
Starting Account Value: $10K, Current Account Value: $30K.
Important: There is no guarantee that these strategies will have the same performance in the future. I use backtests to compare historical strategy performance. Backtests are based on historical data, not real-time data so the results shared are hypothetical, not real. There are no guarantees that this performance will continue in the future. Trading futures is extremely risky. If you trade futures live, be prepared to lose your entire account. I recommend using these strategies in simulated trading until you/we find the holy grail of trade strategy.
We haven’t found the holy grail of automated trade strategy yet, but we get closer with every strategy. Click here for links to all strategy descriptions and the most recent performance chart from the backtest.
Yes! We made it to the first milestone in the Live Test: we’re up $20K to a total account value of $30K! We haven’t reached our goal of $60K, but we’re closer than we were a few weeks ago.
ATS started with a simple nine strategy backtest published in January of 2021. Those backtests led to better strategies that were used to create the ATS Simulated Forward Test. The ATS Simulated Forward Test was started in January of 2023, and the forward test is being used to inform the current ATS Live Test, which started at the end of January, 2024.
Today, I’m going to give you an update on both the ATS Simulated Forward Test and the ATS Live Test.
In the last update for the Live Test the account value was up $8,500, for a total of $18,500. Even though the test officially started at the end of January, it didn’t take any trades until February 20th. Today, the total gain on the Live Test portfolio is 20K.
I’m surprised we got here so fast, especially because it feels like I’ve made so many mistakes. I probably shouldn’t admit that, but I suppose that’s the advantage of automation. It cripples my ability to react in the moment, which requires a great deal more planning and forethought. If I can’t dodge bullets, I must develop ways to avoid them altogether because on the off chance that I do find myself in the line of fire, a shield is my only weapon. With that in mind, it’s hard to proceed in the same way that I would if this were a manual hunt. We’re taking a completely different route through the ‘dark and scary forest’ of trading and it’s unveiling itself to me on a daily basis.
On Day 8 — I was down $2,500 on the day; the account level stop was hit.
On Day 9 — Tuesday of last week, I woke up to a down market and I just knew the Live Test account value would be down as well, but I was saved by a core strategy. I ended the day up $6,500.
On Day 10 — I went to sleep with a negative account value, but when I woke up I felt like this guy:
I can’t watch that enough! I’m going to order a shirt that has ‘ATS’ on the front so I can do this when we hit our end goal.
I’ve learned a lot over the last few weeks and I’m going to use this post to document and share some of the lessons learned. I’m also going to:
show you an updated performance chart of the forward test (NQ portfolio is up $430K from January 1, 2023 to March 7, 2024),
provide a quick overview on the difference between core and non-core strategies; and,
give you a framework for selecting which strategies to enable and how much capital to assign to them.
Let’s get started…
What Is The Holy Grail of Automated Trading Strategies?
Before showing you the best performing strategies in the ATS Forward Test, I want to follow-up with your answers to a survey question in the last Mudder Report. The question was: What do you think the holy grail is?
36% of you said that “it does not exist”
8% of you said that “it is a scalping strategy”
12% of you said that it is a “strategy based on chaos theory”
8% of you said that is is a “machine learning based strategy”; and,
36% of you said that it was “a moving target of multiple algos”.
Clearly, the bulk of you think the holy grail either ‘does not exist’ or is ‘a moving target of multiple strategies’. It’s important to note that these two answers are not mutually exclusive, but if you fall into the latter group, you’re going to love this post because I’m going to give you a framework for a strategy selection process. The first thing we need to talk about is the difference between core and non-core strategies.
Core Vs Non-Core Strategies
If you’ve ever used automated systems to trade live, you know that there are two types of systems (in general)—high risk, high net income strategies, and low risk, low net income strategies. The former is characterized by a high frequency trade count, the latter is low frequency. No doubt the holy grail is an amalgamation of the two, but when using a portfolio of strategies to create the holy grail we need to develop a way to blend both groups. That’s what I’ve been trying to figure out over the last two months and I can’t say that I’ve done it, but I’m certainly getting closer to the mark.
The first group of strategies I refer to as ‘core’; the second group is non-core — easy enough. These definitions could change over time, but in general, the main attributes are a function of return and risk, like every other valuation metric. The hard part is figuring how to value the cash flows. What is the true “cost” of running a strategy? If the cost of capital is a function of risk, how can it be applied to cash flows or ‘strategy earnings’ to get a better understanding of how to rank or value different strategies in a portfolio. This ranking becomes important when trying to decide how much capital to allocate to each strategy.
In general, there are four main attributes or variables when thinking about how to categorize a strategy: net profit, trade count, MAE (intra-trade drawdown), and profit factor. Core strategies tend to have a lower trade count and a lower net profit, which is why I’m using non-core to help boost net profit. Here’s a way to think about where core and non-core strategies fall in relation to net profit and trade count.
In general, non-core strategies have a higher intra-trade drawdown than core strategies. Core strategies also tend to have a higher profit factor than non-core strategies. Here’s a way to think about non-core and core strategies in relation to intra-trade drawdown and profit factor.
In summary:
Core strategies are as close to the holy grail as you can get. They have a high profit factor, low drawdown, and low trade count, but tend to cycle through alpha. The goal is to have 10 to 15 core strategies, but until we get there, the best way to grow account value is with non-core strategies
Non-core strategies have a higher degree of risk, with a higher drawdown, but they tend to have a much higher trade count and net profit.
The selection process for core strategies is rather clear. These are our best strategies: they may not trade often, but when they do trade, I want to allocate more capital than I would to riskier or non-core strategies. The way forward for non-core strategies is less clear so I’ve had to develop a way to think about a way to assign a relative valuation to each strategy. I’m going to share that framework with you, but before going any further, I want to take a minute to review the forward test.
ATS Forward Test Update
We made it to our first milestone in the Live Test, but we’re only half way there. We’ll camp here for a moment as we head into contract expiration next week (triple witching on Friday) and head out again on March 18th.
The next milestone is $45K and the hunt is officially over at $60K or June 24, whichever one comes sooner. The good news is that we’ve learned a lot and I’m going to do what I can to apply that knowledge to the rest of the Live Test. Hopefully, it will make the next summit easier to reach.
With that said, I think it’s time to get into the results of the forward test. I’ll use that to show how I selected which strategies to trade in the live test.
The forward test was started in January of 2023 so the following results are from January 2023 through March 7, 2024. Not all strategies were started at the same time and some are not in use today. The following chart provides you with the NQ-only forward test results.