I think part of the difficulty with college courses is that many students are accustomed to an algorithmic approach to STEM in high school. They are shown problems from a category, shown how to solve them, and then asked to solve a problem from that category on an exam.
An indicator that a student is taking this approach are the questions: “How do you solve this type of problem?” or “What equation do you use for this problem?”
University STEM courses are not taught like that. Part of the reason is that it is not possible to practice every type of problem that a scientist or engineer will encounter in the future. Better questions would be “How do you think about this problem?” or “What concepts are used to solve this problem?”
University science courses tend not to have “the material.” They have ideas and connections between properties/phenomena. So the exams can’t test"the material." What they need to test is the application of the ideas that the students have been seeing, to problems they have not seen before. This is what makes a scientist, after all.
I almost never put a problem of a specific type that the students have seen before on my exams. Having said that, the averages on my exams are usually high 70’s or just touching 80, with an 85 needed for an A. In my opinion, this probably maximizes the work that the students put into the course, and maximizes learning. Lower averages are okay, but I think that a really low average tends to discourage students from trying. I took an organic chemistry class once where one midterm had a median of 23 and a mode of 4.
The British system is different. I think Twoin18 has explained the scoring for various types of honors degrees in math in the UK. In Cambridge mathematics at one time, it was not uncommon for the Senior Wrangler (the top-scoring student on the exams at the end of three years) to have a score that was ten times the score of the student at the far end of the first-class honors degrees (let alone those who got other levels of honors). There was also the general legend that the student who came in second would have a more distinguished career, on the theory that the Senior Wrangler had nothing left to prove, while the person who came second did.
One of my favorite Cambridge math stories is about Philippa Fawcet, who took the math Tripos when women were not allowed to get Cambridge degrees. The practice was to announce all of the men’s results in public (to an assembled group of students), and then for the few women who had taken the exam, to say which men they scored between. In Philippa’s case, the announcement started out “Above,” and the rest of the words, “the Senior Wrangler,” were drowned out in the noise from the crowd.