Find what your trade is quietly assuming.

Last week we made AI argue the other side of your trade. This week, the second of the five questions:

What assumptions am I making?

Every trade idea is a chain of beliefs. Most of them you'd say out loud. A few you never notice, because they feel like facts rather than choices. Those are the ones that do the damage.

"Does my reasoning make sense?"

That question invites a verdict. The AI reads your reasoning, finds it broadly coherent, and tells you so. It usually is coherent. But coherent reasoning built on an untested assumption is still built on an untested assumption.

Stated reasons, unstated premises

Take a hypothetical long on a currency pair after a hawkish central bank statement. The stated reason is simple: the bank sounded tougher than expected, so its currency should strengthen.

Underneath that sit several things you haven't said. That the market hadn't already priced the statement in. That the other currency in the pair isn't about to get its own surprise. That the bank will follow through. That your timeframe is long enough for any of this to matter, and short enough that something else doesn't take over.

None of those is unreasonable. All of them are assumptions. If one breaks, the trade can fail while your stated reason stays perfectly true.

A prompt that does the job

"Here is my trade idea: [instrument, direction, entry, stop, target, and timeframe]. My reasoning: [two or three sentences]. List every assumption this trade depends on, including the ones I haven't stated. For each, tell me whether I can check it now, can only watch for it, or am simply taking it on trust. Then tell me which single assumption, if wrong, would do this trade the most damage."

Three things make it work:

1. It asks for the unstated assumptions explicitly. Left alone, an AI coach will mostly hand your own reasoning back to you in tidier words.
2. It sorts each assumption by what you can do about it, which turns a list into a set of tasks.
3. It asks for a ranking. Assumptions don't carry equal weight, and you need to know which one is holding the trade up.

What to do with the answer

Go through the list once and act on the sorting:

1. Checkable now. Check it before you enter. If it fails, you've learned something for the price of five minutes.
2. Watchable. Write down what you'd see if it started to break, and decide in advance what you'd do. That decision belongs in your plan, not in the moment.
3. Taken on trust. Be honest about how many sit here. One or two is normal. If the most important assumption is in this pile, the trade is mostly faith, and your size should reflect that.

Then record the critical assumption alongside the trade in your journal. When you review it later, the first question is whether that assumption held. It's a better question than what the P&L says, because a trade can win with a broken assumption and lose with a sound one.

You won't remove assumptions from trading. You can stop making them without noticing.

Next week: what evidence currently contradicts my thesis?