Why your practice score is misleading
Practice scores flatter you when the questions are untimed, repeated, drawn from a small bank, or weighted differently from the real paper. A full timed mock on unseen questions, matching the published weights, is the only reliable signal.
A number that says 85 percent feels like readiness. Often it is not, and the reasons are structural rather than a matter of luck.
One: untimed practice
Working without a clock is a different task. You reread, you reconsider, you look things up without noticing.
The real paper is 190 minutes of sustained pace, and an untimed score does not measure it.
Two: seen questions
The second time through a bank you are partly remembering rather than reasoning. The score rises and the knowledge has not.
This is why two mocks should come from different question sets.
Three: a small bank
A bank of 200 questions cannot represent a syllabus this broad. You will score well on the corners of it that happen to be covered and have no information about the rest.
Four: the wrong weights
The real paper is 27 percent origination activities, 24 percent federal law, 20 percent products, 18 percent ethics and 11 percent state content.
Federal law is the most memorable material to write questions about, so practice banks over-represent it. A candidate strong on regulations scores well on such a bank and meets a real paper where three quarters of the questions are something else.
What actually predicts
A full 120-question timed paper, on questions you have not seen, built to the published weights, with no content area below 70 percent.
Two of those above 80 percent is a reliable signal. Everything else is an estimate.
Common questions
Why is my practice score higher than my real score?
Usually because practice was untimed, on seen questions, from a small bank, or weighted differently from the real paper.
What practice score means I am ready?
Above 80 percent on two full timed papers of unseen questions, with no content area below 70.
Why do practice banks over-represent federal law?
It is the easiest material to write questions about, so banks skew towards it while the real paper is 76 percent other content.
Does repeating a question bank help?
For learning, somewhat. For measuring readiness, no, because you begin remembering rather than reasoning.
How big should a question bank be?
Large enough to cover a broad syllabus. A bank of a couple of hundred questions cannot represent this outline.