<p>The reason for the generous curve is that it tests a wide range of math, and the actual subject matter varies a little from test to test.</p>
<p>Their scale is not arbitrary. Who do you think has done more research on it, them or you? If only people with 50/50 got an 800, the score would be a lot more arbitrary, because it would depend on the particular test you get (which is completely random) as much as what you actually know. </p>
<p>There is no useful difference between 49/50 and 50/50, so they should be scored the same. That is not the case with the SAT I. It tests basic math, so a person could be reasonably expected to know everything it tests.</p>