A question asked cold gets answered cold — but a question paired with a guess gets answered against your actual reasoning, and that difference is where most of the learning happens. The First-Guess Rule is simple to state and easy to skip under pressure: before you ask anyone or anything a question, commit to a guess first, even one you're fairly sure is wrong.
Why committing to a guess changes the whole interaction
Think about what a tutor, parent, or AI tool can actually do with a bare question versus a question plus a guess.
A bare "how do I solve this?" gives the answerer almost nothing to work with except the problem itself — they have to guess at your level, your likely misconception, and where you're stuck, all before they can even start being useful. The response ends up generic, because it has to cover every possible gap you might have.
A guess changes that completely. "I think the answer is 15 because I multiplied instead of dividing — is that right?" gives the answerer your exact reasoning to correct. The response can now be precise: it can point at exactly the step where multiplying was the wrong move, rather than re-explaining the whole topic from scratch. You get a better answer because you offered a worse one first.
It makes your reasoning visible — including to yourself
There's a second effect that matters just as much, and it happens before anyone else even responds. The act of forming a guess forces you to notice what you actually think, which is different from what you assume you think. Students frequently discover, in the process of trying to state a guess, that they don't actually have one — which is itself useful information, because it tells you that you're missing something more basic than the question you were about to ask.
This is the general version of what Professor Newton's Predict-Observe-Explain already does specifically for physics: predict the outcome of an experiment before you see the result, then explain the gap between what you predicted and what actually happened. The First-Guess Rule takes that same structure and applies it outside the lab — to a maths problem, an essay question, a history source, anything where you're about to ask someone else to close a gap for you.
A wrong guess teaches more than a right answer received passively
This is the counterintuitive part, and it's worth stating directly: being wrong on purpose, in a stated and specific way, is more useful than being handed the right answer with no guess at all. When your guess is corrected, the correction attaches to a real piece of your own reasoning — you now know not just what the right answer is, but why your particular route to it went wrong, which is exactly the kind of question that sits at the Diagnostic level of the Ask Ladder. A right answer received passively, with no guess preceding it, skips that step entirely — you have the answer, but not the diagnosis of your own thinking.
The three-step routine
You don't need to overthink this in the moment. The routine is short enough to use on almost any question:
- State the question. What exactly are you trying to find out? Be specific — "how do I do this" is too vague to guess against.
- State your best guess and why. Not just an answer — the reasoning behind it, even if it's shaky. "I think it's X because Y" is enough.
- Then ask. Now put the question to whoever or whatever you're asking, with your guess attached.
This works identically whether you're asking a parent, a subject tutor, or an AI tool — see the Ask-Try-Ask cycle for how the First-Guess Rule fits as the opening move of that larger cycle when you're working with an AI specifically.
Where this habit tends to break down
The rule is easiest to skip exactly when it matters most: when you're stuck and frustrated, and the guess itself feels like extra work you don't have time for. That's worth resisting, because a rushed question with no guess attached is the moment you're most likely to get a generic answer that doesn't actually resolve your confusion — meaning you'll be back with the same underlying gap later, just with less patience left to ask about it.
FAQ
What is the First-Guess Rule?
Before asking anyone — a tutor, a parent, an AI — a question, commit to a guess or prediction first, even a wrong one. This converts a passive question into an active one, because the person or tool answering now has something concrete to correct rather than a blank request.
Is the First-Guess Rule only for physics?
No. It's the general version of a technique Professor Newton already uses specifically for physics, called Predict-Observe-Explain. The First-Guess Rule applies it to any subject — maths, English, history, anything — not just experiments.
What if my guess is completely wrong?
That's fine, and often more useful than being right. A wrong guess that gets corrected teaches more than a right answer received passively, because the correction lands on your actual reasoning rather than on a gap you never described.