No Penalty for Wrong Answers

Your score on each section comes from a raw score: a simple count of questions you got right. Nothing gets subtracted for questions you got wrong. That raw count is then converted onto the final scale. Since the formula only ever adds, a wrong answer and a blank answer land you in the exact same place: zero credit. The only way to possibly gain something is to have a guess turn out right.

That one fact flips the logic of blank bubbles. A blank is a guaranteed zero. A guess on a four-choice question is right about one time in four, so filling it in can only help or do nothing. It can never hurt, because there's no penalty to trigger. Mathematically, the expected value of guessing is always greater than or equal to the expected value of skipping, so skipping is never the smarter move.

This connects directly to how the clock works on a linear, non-adaptive test: every question is worth the same, and you see all of them whether you answer early ones right or wrong, so pacing is about coverage, not protecting a streak. That makes the last moments of a section a pure mechanical task. You are not deciding whether to risk anything, because there is no risk left to weigh. The only mistake left on the table is a bubble with nothing filled in.