<p>I’m pretty sure the scale is predetermined based on the difficulty of the test compared to previous tests. As in 99.9% sure. If 2000+s were handed out for a certain percentage of test takers, it would be very difficult to accurately compare two different scores from two different tests. Keep in mind that this is a standardized test and comparing people is its main, if not only, purpose. Ever wonder why they give scaled scores and not raw percentages? </p>
<p>Also, correct me if I’m wrong but I’ve heard from plenty of different sources that the reason CB has experimentals is to determine the difficulty of the section whose scale is not determined yet.</p>