I would love to see one person teaching and a different one grading. Better yet, a panel of credentialed faculty grading assignments to ensure the student learning outcomes are met. This way, the same person teaching does not rate their own student’s learning success. Is the better teacher the one with all the A’s in class or the one with well-distributed grades? We can’t tell by just this information. Then, there is after the course.
How do the students do after your class in another course? That’s more telling than grades. I have students in my quantitative analysis classes that sometimes cannot convert decimals to percentages. I mean, I am stopped in the middle of class when I say there is a 6% probability of xyz (wait, Dr. C. where did the 6% come from?). So, I start with another problem and work it out and I get the same question. Eventually, I realize that I omitted the automatic step of .06 = 6% because to me that is common knowledge.
Does that mean that their algebra teacher failed? What about their elementary school teacher? Well, probably not. These may be students that did not do well in these sorts of subjects to begin with. Maybe, they were barely passed with a C, which means they potentially do not understand 30% of the material (or even more).
The bottom line is that there should be a better system to ensure that all of the required objectives are met by the student and not just 70% of them.