Why are homeworks graded holistically, and tests worth 1 point?

The grading flowchart says that will assign your assignment a score between 0 and 10 points, inclusive. Why are the autograder tests worth only 1 point, and the subjective portion worth the other 9?

So. First off, please recall that we consider a 9 and a 10 /literally/ the same value.

Merely passing the autograder tests is of course insufficient to be a correct solution. We /need/ to manually inspect. Consider the following:

(define (countdown n)
  (cond
    ((0) '(0))
	((5) '(5 4 3 2 1 0))
	(else 'who-knows!)))

Such a definition would pass our unit test suite, but is obviously insufficient. There are also aspects we look for in your solutions that are not expressed in input and output. In fact there are many limitations to using naive automated testing for homework assessments. So instead, we opt for a holistic assessment and use the automated test suites as a guide when evaluating your assignments and also as a minimal-threshold check.