Would this mean that an applicant with 16 points would not have a good chance as an applicant to such a program? I was a bit confused why there are only these 4 clerkships considered - what if the applicant had 5/7 honors but only 16 points due to the HPs falling into these four?
These are descriptive statistics, not prescriptive. These were for applicants who applied to our program the final year I was PD who matched at a "Doximity top 25" program. Yes, my program was/is in that list, but the vast majority of the applicants did not match with us.
By describing the median, I am letting the OP know the level where half of the applicants were at/above the level, with half below the level. Same principle applies for the 25th centile and 5th centile numbers I provided.
Every applicant has a host of metrics to consider, some not easily quantifiable and purely subjective (see my previous posts in another thread about how our program did it--this is just our program, not any other program, each program has its own process). The reason we only recorded 4 instead of being completely detailed and trying to analyze every rotation reflected our approach to sampling, the level of scrutiny and effort it took to figure out the grade distribution of scores for each of those rotations at each medical school and use formulas to adjust each applicant's actual "points" to make them comparable between schools, and the fact that practically all of the applicants had done rotations in these 4 areas as M3 students allowing for comparisons.
Try not to get lost in the details. The OP asked "how important are clinical grades". All I can answer is that applicants who match to "top programs" do well on clinical rotations. How well? Well, usually more honors than not, but not always. Does that mean you have to have all honors to match at a "top program"--obviously not, look at the 25th percentile, about 25% of applicants who end up in a "top program" had 50% honors and 50% high pass for the 4 clerkships (or 75% honors and 25% pass).
None of these metrics are independent. Medical students who do well on USMLE exams often do well on clinical clerkships, often have higher number of publications, often have better MSPE "rankings". So, using our method, does the student who has 16 points have less chance than the student who has 20 points?--maybe, but not because of the clinical grade difference at all, but rather because the student with 20 points is more likely to also have higher USMLE scores, higher MSPE rankings, higher # of publications--and, truth be told, more likely to have a better interview. We used the clinical grades as part of multiple metrics to figure out who to interview--but I don't actually ever recall ever using clinical grades to distinguish how we were going to relatively rank two candidates AFTER we interviewed them.