If you haven't found the holy grail of automated trade strategy yet, why are you testing live?

We're in a state of constant refining and iteration

As a quick reminder, we’ve defined the holy grail of automated trade strategy to be the following:

• Profit factor greater than 3

• Annual drawdown less than 3%

• Annual return on max drawdown greater than 500%

• Maximum cumulative daily low of -$1,000

• Avg Daily profit greater than $1,000

• Less than 5,000 trades annually

• More than 253 trades annually

We haven’t found the holy grail of trade strategy just yet, but we’re looking for it and part of the hunt includes live tests of our best strategies to make sure we’re on the right track.

I’ve been asked the following question several times over the last week and I think it deserves a public answer:

If you haven't found the holy grail of automated trade strategy yet, why are you testing it live?

I’m not sure what’s triggered folks (perhaps the funding piece), but this was the plan all along. What’s more, I’m not sure how it hurts or hinders the hunt to conduct a live test at any point in the process.

The goal is to start the live test in January of 2023. We’ve developed over 50 automated strategies and now we’re testing them. We’re testing them for performance and backtest accuracy. Part of that test is a live test. The first live test will be with a $50K account. It uses real dollars so I want to exhaust our other options first. Our other options include a real-time test on a virtual server as well as an evaluation test for funding. The evaluation serves as both a test for our hunt as well as the possibility to secure funding for the live test in January.

Why test strategies that don’t yet meet the criteria for the holy grail?

There’s preliminary evidence to suggest we’re on the right track with our new optimization strategy. The strategies are the same, but the time series has changed to something that both improves backtest accuracy and strategy performance. It will take some time to see if these new strategies will continue to produce the same profit factor over time.

That said, if they do, many of the portfolios we’ve created with these strategies surpass the attributes we originally required for the holy grail of automated trade strategy, especially with regard to profit factor and drawdown. You can read more about what we found in Portfolio Strategy 54, where we developed 6 different portfolios from a master list of over 300 strategy variations.

Portfolio 54a is my favorite portfolio. It consists of only four strategies and has a total net profit of $626K for the year. The profit factor is low at 1.44, but the net profitability makes it worthwhile and there might be opportunities to improve profit factor by focusing on certain days of the week.

At the same time that I’m testing the performance of strategies, in particular the new time series and portfolio theories as described above, I’m also testing the accuracy of the backtest to real-time trades. My hope is that by the beginning of December we’ll have enough evidence to be able to trust in the accuracy of the backtest and the consistent performance of the new portfolio.

Subscribers can follow or track our process every week in the Mudder Report.

If we do move forward, I want to have a few sources of funding lined up to conduct the first live test. The degree of certainty that we have around the performance of the strategy will dictate the profit split. The test or evaluation with the funding partner is also to help prepare for the logistics of setting the strategy up on a live account (with or without a funding partner).

I am also considering setting this up as a test on GoFundMe via donations to mitigate any legal concerns involved with testing investment products. It would also eliminate any profit split or drawdown rules, but I’m not sure if it’s feasible without a social tie-in like “if the strategy is successful, we’ll pay everyone back and use the first $100K to fund ‘GoFundMe’s’ related to health care.”

I welcome your input and ideas on the best ways to move forward on this as well as the test in general. My motives are not purely financial so I also value socially motivated ideas.

The next Mudder Report will be published on Sunday, September 25 and Portfolio Strategy 55 will be published on October 1. Portfolio Strategy 55 is like Portfolio Strategy 54, except it focuses on micro contracts.