Sorry but this makes no sense. There are far more applicants who can pass the courses and graduate (“succeed” by your definition) than there are places available at a college like Berkeley. This is not an open access community college.
So there has to be some basis for selecting between them. The (holistic) academic ratings appear to be about estimating the future level of success of prospective students. So (simplifying hugely) a prospective student who has the potential to get a 4.0 GPA in college should be rated more highly than one who is only thought like to get a 2.5 GPA, even thought that latter score is “success” in terms of gaining a degree.
Inevitably such predictions about future success are often inaccurate, but that doesn’t imply that they “mean absolutely nothing”. If I hire someone to do a job, then I’m trying to determine who I think will be most successful in that job. I may be wrong, but it would be crazy not to try and exercise some judgment in choosing between applicants. I’m certainly not going to conduct a lottery amongst everyone who could do the job.