Hawai'i Trading Academy | Trading Coaching & Education

"If the tests pass, the tool works." That's the assumption almost everyone makes about AI-built software. We made it too. For about twenty minutes.

Earlier this year we had Claude build us a position-sizing guardrail tool. Something to catch us before we sized a trade too big. The first version came back looking sharp. Every test passed. The documentation read like a senior engineer wrote it on a good day. Our gut said ship it.

We didn't. And that decision is the whole point of this post.

What Was A "Perfect" Build Actually Hiding?

I (Reid) run point on our AI builds, so I was the one staring at that first version, ready to call it done. Then we did what we tell every student to do with a new strategy before it touches real money. We audited it instead of trusting it.

What we found wasn't a small bug. It was three of them, stacked underneath a shiny surface.

The tool was reading from a dead data file. A source that no longer existed in the pipeline it was supposedly checking. Run...

Our position-sizing tool sat quiet for 30 trades before we let it touch anything. It watched real trades come in. It logged what it would have told us to do. It changed nothing.

That wasn’t caution for caution’s sake. That was the plan from day one.

This is post one in a ten-part series on what we’ve actually learned building tools and testing strategies with Claude, Anthropic’s AI. Glenn and I aren’t AI developers by trade. We’re traders who started using Claude to build things we needed and couldn’t buy off the shelf. Some of what we built worked. Some of it didn’t, until we fixed how we were building it. This series is the honest version of that process. What to do, what not to do, no polish added.

Today’s post covers the do’s. Specifically, the three things that kept an AI-built risk tool from ever putting our account in danger, even while it was still rough around the edges.

What’s the One Rule That Comes Before Any Code?

I (Reid) handle most of the AI and content systems at ...

Most trading content answers the wrong question. It tells you what should happen next. Good NQ futures research starts somewhere else: what actually happened the last time this condition showed up, how often did it happen, and how wide was the range of outcomes?

That difference sounds small, but it changes the way you prepare. A chart opinion gives you a story. Research gives you a distribution, a sample size, a definition, and a reason to know when the story is too weak to trust.

Why HTA Is Building an NQ Futures Research Lab

At Hawai'i Trading Academy, our core pillars have always been Risk Management, Edge & Strategy, and Psychology & Process. Research sits right in the middle of all three. It helps you test whether an edge is real, keeps risk expectations grounded, and gives your brain something better than recency bias to lean on when the market gets loud.

We already teach traders to backtest before they trust a setup. In our guide on using TradeZella for backtesting, the poin...

A trading statistic can be technically correct and still lead you to a bad decision. That is the problem with screenshots that say '72% win rate' and stop there. The number might be real, but without the sample, condition, distribution, and downside, you do not know what it actually means.

If the HTA Research Lab is going to be useful, traders need to know how to read NQ futures statistics without turning historical probabilities into predictions. Here is the framework we use.

1. Start With the Definition, Not the Percentage

Before you look at a result, define the event. 'Gap fill' sounds obvious until two traders use different closes, different opens, different sessions, and different thresholds. 'Trend day' is even worse. One person means close above open. Another means one-directional price action with shallow pullbacks. Those are different studies.

A clean research page should tell you exactly how the condition was measured. For NQ, session boundaries matter. So do the timefram...

Somewhere between 80 and 100 prop firms disappeared in 2024. Not “struggled.” Gone. If you were shopping for a funded account two years ago, a big chunk of the names on your list don’t exist anymore.

That sounds like bad news. We think it’s the opposite.

From roughly 2020 to 2023, the prop-firm world ran on a gold-rush script: cheap challenges, easy funding, big promises, a new firm launching every week. 2026 looks different. The industry now calls it the “operator era” — fewer firms, higher standards, and real weight on trust, risk control, and education. The market quietly repriced hype.

We watched the whole cycle from the coaching side here at Hawai‘i Trading Academy. And the consolidation confirms the thing we built this place around: durable skill beats hype every single time the tide goes out.

What actually happened to all those firms?

Two things, mostly. First, a lot of them were undercapitalized. They sold cheap challenges and paid out on the accounts that passed, quietly ...

Today we are opening the HTA Research Lab with one simple idea: stop guessing and ask the data.

There is no shortage of futures opinions online. There are calls, predictions, screenshots, hot takes, and clean explanations written after the move already happened. What traders need more of is transparent futures trading research that shows the sample, the condition, the result, and the limitation in the same place.

What the HTA Research Lab Is

The Research Lab is a public market-research experience built around questions traders actually ask about NQ futures. Instead of opening a page and being told what to trade, you choose the market context you care about and inspect what happened historically under a frozen definition.

The MVP starts with NQ only. That is intentional. We would rather launch a smaller set of clear, useful studies than dump 100 research cards into an interface and make traders hunt for the point.

You can explore five main timeframes: 5-minute, 15-minute, 1-hour, 4...

Max Loss Days: Setting and Enforcing Hard Stops

Every professional trader has a max daily loss. Most retail traders don't. This gap is why one group makes money and the other bleeds it.

A max daily loss is simple: you decide in advance, while your head is clear, what the largest loss you can take in a single day looks like. Then you enforce it. No negotiation. No exceptions.

Setting One That Actually Works

Your max daily loss should be based on your account size and your strategy's expected drawdown profile. A common starting point: 2% of your account. On a $50K account, that's $1,000. On a $100K account, $2,000.

But the 2% figure is a starting point, not gospel. Some strategies with higher win rates and lower average losses can handle 3%. Some volatile strategies need 1% or less. The key: it should be large enough that you can take 2-3 normal losses without hitting it, but small enough that hitting it doesn't put your account in jeopardy.

Why Enforcement Beats Willpower

Setting...

Tropical Storm Lala is heading straight for the islands this weekend. Forecasters have Hawaiʻi Island under a hurricane warning, with damaging wind, dangerous surf, and up to a foot or more of rain spreading across the state from Friday into Sunday. If you trade from Hawaiʻi, this is a live risk-management drill whether you asked for one or not. So here is the Hawaiʻi Trading Academy storm-day plan: how to protect your account, and your head, when the weather takes the decision out of your hands.

Why a storm is a risk problem before it's a weather problem

Trading is already the practice of managing what you can't control. A hurricane just says it louder. You can't control Lala's track, the rain totals, or whether your neighborhood keeps power Saturday night. What you can control is your exposure before any of that happens.

The mistake is treating a storm week like a normal week with a little extra weather. It isn't. Power flickers. Internet drops. Cell towers get congested right whe...

Correlation Risk: Why Your Diversified Trades Aren't

You have three trades on: NQ long, AAPL calls, and a TQQQ position. You think you're diversified. You're not. You have one trade on, three times.

This is correlation risk. It's the invisible killer that turns a manageable losing day into a catastrophic one.

What Correlation Risk Actually Is

Correlation measures how closely two instruments move together. A correlation of 1.0 means they move in perfect lockstep. A correlation of 0 means they're independent. A correlation of -1.0 means they move in opposite directions.

NQ and ES? Correlation typically sits around 0.92-0.97. They're basically the same trade. NQ and AAPL? Around 0.85. NQ and TQQQ? Around 0.98. If you're long all three, you don't have three positions. You have one position, three times the size.

Why This Matters for Futures Traders

Say your risk model allows 2% total account risk at any given time. You put on NQ at 1% risk and ES at 1% risk. Your model says you're a...

The economy lost 23,000 jobs in July. Wall Street threw a party. The S&P 500 closed at a record 7,757, the Nasdaq jumped about 1.3%, and NQ futures ran up roughly 1.2% on the day.

If that makes no sense to you, good. It means you are paying attention. A shrinking job market should scare investors. Instead, it thrilled them. So why did stocks rip on obviously bad news?

The answer is the single most useful thing a new trader can learn about how markets actually work. It is not the number that moves price. It is what the number does to the Fed.

Wait, the economy shrank and stocks went up?

Let's set the table. Economists expected around 83,000 new jobs in July. Instead, payrolls fell by 23,000, and prior months were revised down hard. On the surface, that is a soft labor market flashing a warning.

The day before, futures markets put the odds of a September rate hike near 55%. Within minutes of the report, those odds collapsed toward zero. Traders decided the Fed now has cover to leave...