<p>No. Here is how it is done:</p>
<p><a href=“College Board - SAT, AP, College Search and Admission Tools”>College Board - SAT, AP, College Search and Admission Tools;
<p>The percentile ranks DO change from one year to another (see above). I.e., a score of 700 may be the 93rd percentile in one year but the 91st percentile the next.</p>
<p>Here is a hypothetical:</p>
<p>An SAT is given in March with some curve. The exact test is given again in June. Suppose a bunch of very good math students take only the June test. Two very similar average people take the test, one in March and one in June, and both get the same number of questions right.</p>
<p>In the actual scoring method, each person would get the same score (say, 500), since the curve is unchanged by the appearance of the good math students.</p>
<p>In your method, the March person gets 500, and the June person scores, say, 480. But the ability of each student is identical. This would be very bad from the CB point of view: they want a given score to mean a fixed ability level from one test to another, and from one year to another, etc. Otherwise, the SAT is pretty useless for college admission purposes.</p>
<p>I suppose you could believe the above CB document is fakery. But why would they make their own test scores mean one thing one month, and another thing a different month?</p>