How the LSAT is scored: raw score, equating, and the band nobody mentions

The LSAT is scored on a scale from 120 to 180. Everything interesting happens between the questions you answer and that number, and the parts nobody explains are the parts that matter when you are deciding whether to sit again.

Step one: the raw score

Your raw score is the number of questions you answered correctly across the scored sections. LSAC's specifications describe a test of four 35-minute multiple-choice sections — one Reading Comprehension, two Logical Reasoning, and one unscored variable section. Only the three scored sections feed the raw score.

The unscored section is real and it is invisible. It can be either Logical Reasoning or Reading Comprehension, and it can appear at any point in the test, so there is no way to identify it while you are sitting there and no benefit in trying. People who spend the exam guessing which section does not count are spending attention they need elsewhere.

There is no penalty for a wrong answer

LSAC states it plainly: there is no deduction for incorrect answers. This has one immediate practical consequence that a surprising number of test takers get wrong — never leave a bubble empty. A blank and a wrong answer cost exactly the same, so a blank is a guess you declined to make.

It also changes how you should handle a section running out. With two minutes left and six questions unread, the correct play is to answer all six and then work whichever you can, not to work two carefully and leave four blank.

Step two: equating

Raw scores are not directly comparable across test forms, because forms differ slightly in item difficulty and in the number of items. LSAC uses a statistical procedure called equating to make scaled scores comparable across forms.

In plain terms: if your form ran slightly harder, the conversion accounts for it. This is why the number of questions you can miss and still reach a given score is not fixed, and why anyone quoting a universal figure is quoting one specific released form.

It also means your score is a statement about your ability relative to the equated scale, not about that day's answer key. That is the whole point of the procedure, and it is the reason a score from one administration can sit next to a score from another on an application.

The score band, and why it should change your retake math

This is the part almost nobody brings up, and it is the most decision-relevant thing on this page.

LSAT scores are reported to law schools with a score band, because the estimate of proficiency a single score provides is not perfect. The band is derived from the standard error of measurement, which LSAC describes as very stable and about 2.6 scaled score points.

So the school does not just see 164. It sees 164 alongside a range that acknowledges your actual proficiency might sit a little above or below it. Two applicants three points apart are, in measurement terms, not clearly distinguishable.

Apply that to your own practice tests. A 161 followed by a 164 is not evidence that you improved. It is well inside the noise a stable test produces on a stable ability. A trend across five sittings means something; the gap between two adjacent ones usually does not, and reading progress into it is how people talk themselves into and out of retakes for no reason.

What this page will not tell you

Percentile figures are likewise left out here. LSAC publishes percentile tables covering specific testing years, and a percentile quoted from the wrong span is worse than none — it is precise and wrong. Check LSAC's current table directly when you need one.