Paste in a universe of stocks and ETFs. Strategy Mill puts Portfolio Boss to work on each one, runs the machine-learning search, filters the failures, and brings you back to the small group worth your attention.
Now you know what Strategy Mill does: it searches a much larger universe, throws away weak candidates, and brings you back to the survivors. But finding a beautiful backtest is not the finish line. The question that matters is what happens after you stop touching the strategy.
So in July 2026, we turned that question into a public experiment: the $10,000 Portfolio Boss AI Trading Challenge. Entrants build their portfolios, submit them, and then the files are timestamped and locked against edits. No do-overs. No hindsight. From that moment forward, only new market data gets a vote.
| Rank | Entrant | Strategy | PB Score | Return | Max DD |
|---|
These models come from some of the best-funded AI laboratories on Earth. I gave each one the same unpleasant rule every human entrant faces: build a ten-strategy portfolio, lock it, and live with whatever the market does next.
No mid-course rewriting. No quietly replacing a bad strategy. No peeking at the answer and pretending the machine knew it in advance.
The four model entries are exhibition participants and are not eligible for Challenge prize money.
The Challenge is open through December 31, 2026. Portfolio Boss Evo gives you a free entry path. Strategy Mill is optional—but it can automate the most repetitive part of building a serious multi-strategy entry.
The clock matters. Your score does not begin until your portfolio is submitted, accepted, timestamped and locked. The sooner you finish the research, the sooner your strategy gets a chance to compete.
Help Me Build My Challenge Entry Faster →The machine-learning search is not the painful part. Repeating the setup, launch and review process across an entire market universe is where the project quietly turns into a second job.
Choose the settings, launch a 100-generation search for one stock, wait for it to finish, and then start over with the next symbol.
This demonstration compresses one search into about nine seconds. A real research run takes much longer.
Move the slider. The estimate assumes only three minutes of setup and first-pass review per symbol—a very charitable assumption.
Watch a large universe enter, pass through machine-learning search and filtering, and emerge as a much smaller shortlist. The goal is not to force a strategy out of every stock. The goal is to reject aggressively.
Interactive illustration. An actual large-universe Strategy Mill run can keep your computer working continuously for several days, depending on the number of symbols, test settings and speed of the PC. Counts and sample symbols shown here are not promised outcomes.
The highest return is not automatically the strongest candidate. Pick the strategy you would advance to the next round.
Illustrative metrics for explaining the selection problem; not actual performance results.
I did not start out with a $6.2 million machine-learning platform. I started trading in 1997, blew out three trading accounts, and finally got fed up enough to make myself a promise: never place another trade I could not test with a computer first.
That obsession with evidence eventually became The Relaxed Investor, which went on to more than 500,000 downloads. Later I ran Million Dollar Target publicly, with the results audited by an independent CPA. Instead of taking the win and moving on, I rolled the money back into the next problem I wanted to solve.
Portfolio Boss 1.0 launched in 2014. Since then I have poured more than $6.2 million into building machine-learning tools that can reverse-engineer objective trading rules for stocks and ETFs.
And after all of that, I ran into a bottleneck so stupid it was almost funny: the computer could do the sophisticated work, but I still had to feed it symbols one at a time. Prepare one. Start one. Wait. Review it. Then do the next.
That matters because useful patterns are not everywhere. The needle is rare, and the haystack is enormous. If only a small fraction of symbols contain something worth keeping, the answer is not to stare harder at charts. It is to search vastly more hay.
Strategy Mill became that missing conveyor belt. You copy a large symbol list, paste it into Portfolio Boss, press Start, and let the system move through the universe while you do something else. It builds candidates, applies the chosen filters, rejects weak output and brings you back to a much smaller group worthy of human attention.
It is not designed to manufacture a glamorous strategy for every ticker. It is designed to say no hundreds of times without getting tired, bored or emotionally attached to the answer.
Paste a list of stocks and ETFs into the workflow. No programming is required.
Portfolio Boss builds and tests machine-generated rules across the list.
Weak candidates fail the criteria instead of consuming your review time.
Examine the strongest candidates and decide what deserves a place in a multi-strategy lineup.
Decades ago, a hedge fund manager taught me that consistency rarely comes from finding one heroic system and asking it to carry the whole account. It comes from combining several sound strategies that do not all stumble at the same time.
Drag the white divider below. Both sides end at the same hypothetical 46% annualized growth path. The difference is the ride.
The left side concentrates the entire journey in one return stream. The right side combines ten independently behaving strategies and averages away more of the short-term lurching.
This is why Strategy Mill promises more than saved time. It makes a larger search practical, and a larger search gives you more raw material from which to build a genuinely multi-strategy portfolio.
Instead of spending your limited time manufacturing tests, you spend it judging evidence.
The Challenge shows what happened after the rules were locked. These statements show that strategies discovered through the Strategy Mill research workflow have also crossed the line from computer research into live brokerage accounts.
Russell's account gained 3.50% during September 2025. The dollar result is unusually large because his account is unusually large.
Individual result. Not typical. Click the image to enlarge.John's account statement shows a $36,028.86 increase and a 6.39% period return over roughly one month.
Individual result. Not typical. Click the image to enlarge.Yvonne, then a new trader, recorded a 10.36% period return from April 1 through May 2, 2025, during a particularly volatile stretch.
Individual result. Not typical. Click the image to enlarge.Again and again, the same themes show up: more systematic trading, less guessing, stronger multi-strategy thinking, and tools that let the computer do more of the repetitive work.
“This lets the everyday trader compete with the big boys.”
“Having a mechanical system makes life a whole lot easier.”
“This is a simple process that allows me to get on with my day.”
“Combining several of the strategies together [can] produce even smoother returns.”
“Trading as a business… having the correct tools makes all the difference.”
“What a sense of empowerment! Every day now, I power up ‘The Boss’ and see what edges it can find for me.”
“His structured, systematic, data driven system is proven… Dan strives to find edges.”
Testimonials reflect individual experiences. They are not a promise of results. Trading involves risk, and hypothetical or simulated results have limitations.
I lowered the price because I wanted a broad cross-section of Portfolio Boss members—including people with very little machine-learning experience—to stop watching from the sidelines and enter the $10,000 AI Trading Challenge.
That price is not going to last forever. The Challenge was the reason to make Strategy Mill almost absurdly accessible while people still have time to build, lock and forward-test an entry before December 31, 2026.
Portfolio Boss Evo gives you the machine-learning engine. Strategy Mill gives that engine scale.
Use Strategy Mill for 30 full days. Give it a real symbol list and see what changes when the repetitive work moves from your desk to your computer.
If Strategy Mill does not make that process materially easier for you, contact my staff within 30 days. We will promptly give you a full refund.
No. Strategy Mill is designed around a copy-and-paste workflow. Portfolio Boss performs the machine-learning work; you choose the process, launch the search and review what survives.
A large-universe run can take several days of continuous processing, depending on your computer, the number of symbols and the test settings. That is the point: your PC can keep working 24 hours a day without requiring you to launch the next symbol manually.
No—and that is not the goal. A useful discovery system needs permission to reject most candidates. Strategy Mill helps you search a much larger field and narrow it to the candidates that meet your chosen criteria.
Yes. Strategy Mill removes repetitive setup and much of the initial filtering; it does not replace human review. You still inspect the survivors, verify the results and decide what deserves further testing or inclusion in a multi-strategy lineup.
Yes. Portfolio Boss Evo gives you a no-purchase entry path. The Challenge runs through December 31, 2026, with $10,000 for first place, $2,500 for second and $1,000 for third. Strategy Mill is optional, but it can make the research process much faster to launch and much easier to scale.
Your Challenge score begins only after the strategy file is submitted, accepted, timestamped and locked. A finished portfolio sitting on your computer cannot accumulate forward results. Earlier completion gives your entry more market days before the scoring window ends.
The Challenge is a locked, forward-tested strategy competition. The rules are frozen before later market data arrives, but the leaderboard is not the same thing as a brokerage statement. The separate brokerage examples on this page are real-money account statements.
The regular price is $399. I reduced it during the Challenge because I wanted a much broader group of Portfolio Boss members—including inexperienced traders—to build serious entries instead of leaving the software untouched. This special price is not intended to last forever.
Use it for 30 days. If it does not make large-universe strategy research materially easier, contact Portfolio Boss support within that period and receive a full refund.
I used to program every idea, stare at charts, dream up the next thing to test, run it, reject it, and start again. Finding enough different strategies to build a serious portfolio could consume months of work. Some parts of the process took years to develop.
Now you can copy a symbol universe, paste it into Strategy Mill, press Start, let the computer crunch numbers around the clock for a few days, and come back to the filtered survivors.
That is a mind-boggling change in what an individual trader can do from an ordinary PC.
I created the $13,500 AI Trading Challenge as a dangling carrot to get people off the sidelines and actually use these tools. First place wins $10,000. Second wins $2,500. Third wins $1,000. Entry is free, and scoring continues through December 31, 2026.
The sooner you build and lock your portfolio, the sooner it gets a fighting chance to accumulate forward results. There is no reason to keep watching other people do it.
Strategy Mill normally sells for $399. During this special Challenge offer, it is $39.
Dan MurphyCopy the universe. Start the search. Come back to the survivors.
Discount Applied Successfully!
Your savings have been added to the cart.
Please confirm you want to block this member.
You will no longer be able to:
Please note: This action will also remove this member from your connections and send a report to the site admin. Please allow a few minutes for this process to complete.
