<p>Isn’t that the answer, though. Almost by definition, deferred SCEA applications will do somewhat better than the average for the RD pool, because X% of the RD pool never had a meaningful chance and is going to be rejected more or less automatically, and none of the deferred SCEA applicants are in that group (if they were they would have been rejected EA). Assuming that X% is some meaningful percentage (20%? 30%? maybe even 50%), deferred SCEA applicants would have to do a LOT worse than the serious-candidate RD pool to do as badly as average for the whole RD pool. By the same token, deferred SCEA applicants probably don’t do as well, on average as the serious-candidate RD pool, because that pool includes Y% of superstar applicants who would have been accepted in a jif at the EA stage had they applied then. </p>
<p>So, my guess is that deferred SCEA people do better than the RD average, but worse than the “real”, serious-candidate RD average. Whether they do a little better or a little worse than the good-not-superstar portion of the RD pool, who knows? It isn’t really worth fretting about.</p>