<p>
</p>
<p>I think this is because, on average, the people who get 2400s on the SATs fall into one of two categories. Either they are just naturally really smart and got a 2400 on their first try, or they got a 2100 (or so) and then studied like crazy until they got a 2400 (after possibly multiple retests).
People who fall into the first category probably have great grades to back up the 2400, and likely also have some other sort of academic awards. So they’re accepted for the whole package, not just the 2400.
People who fall into the second category are probably very hard working and dedicated to going to a top college (to study so hard just to bring up an already very good score). Thus they probably devote the same effort to their high school grades and have spectacular GPAs, and also devote time to ECs to make sure to get into that top school. So they, too, have the full package and not just the 2400.
On the other hand, you can also divide those who get ~2100 into two rough categories: those who studied hard (or didn’t), got the 2100, realized it was a great score, and then decided to spend their time doing more worthwhile things than retaking the test. If they also have great GPAs and ECs, they’ll get into top schools.
There’s also the group who studied hard, got a 2100, and said “wow, that’s more than I thought I’d get. Great!” and then continued to waste their time on other unproductive things. If they don’t also have great GPAs and ECs, they wont be accepted, even if they do have the 2100 (which I believe is a great score).</p>
<p>tl;dr - Correlation doesn’t always imply causation. Just something to think about.</p>