Automated Trading Strategy: X

It's time for another live test!

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.

---

Quick note: There is currently a one month wait on joining ATS (free subscription). It is my sole discretion to remove any subscriber.

---

Photo by Towfiqu barbhuiya on Unsplash

We’re on a hunt to find the holy grail of automated trading strategy. We haven’t found it yet, but this strategy is the closest we’ve ever been.

What does that mean?

Well, three years ago we started with nine strategies that had a backtested profit factor of 1.10. Oh, I really felt like we were doing something.

Then we added more strategies and more backtests. I remember how good it felt to find a strategy with a backtested profit factor of 1.50. Oh, I really felt like we were doing something.

Then we tested more methods, developed more ideas until we had over 60 strategies. Each of which has its own advantages and disadvantages. One in particular (56c) performed so well in the backtest, that I used it to apply for funding. It got funded! Oh, I really felt like we were doing something.

Unfortunately, it hit a high of $12K and then blew up due to some rather stringent rules.

But we’re learning from all the mistakes. And, I was serious when I said my goal is to never reach 100 strategies.

Today, we have a list of 70+ backtested strategies. The average profit factor of the backtested portfolio is 2.19; many strategies have a higher profit factor than that. I’ve even published a few strategies with a backtested win rate of 100%. Stats like that would have blown my mind when we first started and it’s a good first step, but backtests are flawed in two ways:

1. historical data is often flawed (garbage in)

2. the simulation is flawed (garbage out)

Don’t get me wrong, backtests are a good first step, however, they can only point us in the right direction.

In January of 2023, after a few failed forward tests in September of 2022, we started the current forward test with around 8 strategies. Today, the forward test has well over 35 strategies and over 130 different variations on those strategies. Variations differ based on:

• data series (minute, volume, tick),

• bar formation frequency (1 minute, 2 minute, 30 minute),

• backtest type (general optimization vs genetic algo),

• risk management,

• instrument type (NQ, MNQ, ES, etc); and,

• a host of other attributes.

While there are several strategies in the forward test that are performing well, especially on certain instruments, none are doing as well as Strategy X.

To be clear, a forward test is a test that uses live data to test strategies on a simulated account. Instead of testing a strategy using historical data on a simulation, we’re running a strategy using live data, in real-time. The only thing that is not real is the account itself. So the results are still hypothetical, but we’re beyond the veil.

What is Strategy X?

Strategy X has already set a record for the number of consecutive win days in a row for any strategy in the forward test. I started forward testing this strategy on 10/26 and it has been profitable on every day since.

• Three days in a row is good, but it happens all the time, especially when market conditions are less volatile.

• Five days in a row is great, but possible. Many strategies perform well in one week and poorly in the next.

• Seven days warrants my attention. Suddenly, this strategy is wearing a pair of Manolo Blahniks.

We’re at 11 days now and the strategy has been profitable every day. This has become THE strategy to beat.

I’ve been doing this for long enough to know that two weeks of consecutive positive net profit means nothing, but it’s a great start. Especially when you consider:

1. that the data is based on two forward tests, not backtests.

2. performance has been verified based on two different data sources, three computers and several instruments.

3. the market has been crazy over the last two weeks: we’ve had earnings reports, Fed Chair speeches, FOMC rate announcements and world events that appear to grow more volatile by the day.

Before you get too excited, let me tell you that I won’t be sharing how to recreate this strategy with you today because:

• We want to study its performance for a few months before sharing

• I want to keep the experiment as clean as possible

• I want to see if there’s a change in performance after sharing

Of course this is all predicated on the strategy’s continued good performance.

I’ll be sharing the performance of Strategy X with everyone today and in the Mudder Report. If it continues to perform well, I promise to share how to recreate the strategy with you around March or April of 2024.

What can I tell you about Strategy X?

What I can tell you about Strategy X is that it is a variation on a strategy I’ve already published. About three weeks ago I received the variation from a subscriber who does not want acknowledgement. He made a slight change to the strategy and got some really amazing forward test results. I started forward testing the strategy myself about two weeks ago. These are the results:

To be clear, these are not backtest results. These are results from the trade report, which is based on live trades.

Here’s an overview on a daily basis:

This strategy was only started on 10/26. It is trading with one contract on three different instruments. Together, over 127 trades have been made, but the win rate is higher than 87%.

I’m going to take the next few months to study this strategy. What makes it so profitable (at least for now)? Is it possible to improve it?

Why the name ‘Strategy X’?

I’ve decided to name the Strategy X as an ode to Elon Musk—I’m joking. This a variation on another strategy I’ve already posted so I don’t think it counts as a new strategy…yet. I’ll be posting a new strategy shortly that will be Strategy 75 and Strategy X won’t become its own strategy until it is ‘properly’ published.

My Challenge To You

Photo by Eden Constantino on Unsplash

We’re swimming in a stew of uncertainty, but one thing is certain: the most important tool in this hunt is the forward test. I really can’t stress this enough.

So here’s my challenge to you: if you haven’t done so already, create your own forward test. Just do it. Even if it’s only one strategy, make it happen today. With or without ATS, the more strategies you track, the closer you’ll be to finding your own holy grail. Most importantly, it gets rid of the need to validate historical data. Half of the hunt is just making sure the backtest data is clean and the backtest/optimization itself is reliable. You can eliminate these two giant obstacles overnight just by starting your own forward test.

Here’s my promise to you for a little added inspiration: If you do find a strategy that has a positive net profit for 10 days in a row, send me the results. If I get the same results over the next 10 day period, I’ll not only share Strategy X with you, but I’ll comp your subscription fee for the duration of the service. It is completely up to you if you want to share the strategy with the rest of the group or not.

Which strategies should you use in the forward test?

Start with the one that you would trade live today if forced. Then add a new one every day until you have at least 10 strategies going. You can run the strategies that I’m running in our forward test. You can run your own strategies. You can run many different variations of the same strategy. That is, if you only have one strategy, run it using a different data series and/or backtest methodology and make a hypothesis about the impact of the change. Use the outcome of that test to create your next strategy or variation, and so on.

Another benefit to running your own forward test is that you get to monitor the activity of strategies in real time. This is something I only started doing recently, but I wish I’d started sooner. In this way, certain strategies have become like indicators. I’m starting to get an understanding for which strategies are correlated and/or more active in certain markets over others.

What’s Coming In November

I had to take a week off last week, so we’ve got a lot on deck in November:

• AI update: New AI generated strategies and an update on how the AI generated strategies we started running in the forward test are doing.

• New HWR strategy

• Forward test update

• Help For Those Trying To Become a Funded Trader: My pro tips on how to pass an evaluation.

• Update on how Strategy X is doing in funding evaluation: Starting Monday, we’ll be running Strategy X on a $100K evaluation to see how it performs.

The hunt continues…

If you have any questions, please let me know or comment below. If you reach out to me on chat, there’s a chance I won’t see it. If I have yet to respond to your email/chat, pls resend.

Contact: AutomatedTradingStrategies@protonmail.com.