A teacher who can explain difficult concepts can save you literally dozens of hours of self-study. At least, this has been the case in my experience. Having someone to ask questions of is huge. Having a good study group is not always a dependable solution.
On a more immediate level, if I have to spend three or four hours of useless time in class, that's time taken out of my schedule to actually learn the material.
Granted, I have a commute and a job and family, so it's not quite the same as if I were a 20-something undergrad, but should we value their time less than my own? Many universities are chock full of teachers who are good researchers and grant writers, not necessarily good professors, and I do feel that this is a big problem.
Also, Teaching Assistants really should be offered more in every level of education. A good TA can reduce the instructor's load significantly, while at the same time offering much better interaction ratios with students with questions.
Sure it did, but poor teaching was not the reason, the difficulty was. My university has special courses on most fundamental topics, the difficulty of which is ridiculously high compared to regular ones (which are much more difficult than on most other schools in my country anyway). Although these are taken mostly by really great minds, completing them with good grade is next to impossible without spending a lot of time on research, discussion with friends and instructors, and problem solving. Having that done, problems on regular exams seem trivial, and possibility of someone not being able to solve them is unimaginable.
These words remind me of my introductory class in graduate solid state physics. The instructor was teaching the course for the first time, or at least the first time in a while. His lectures were very difficult to follow, the notes we took weren't of much use, and he hadn't calibrated the problem sets quite right. So, omigod, the problem sets were difficult. They were due weekly, they took two or three days each, and usually at least one all-nighter.
It was hell. But in retrospect it turned out amazingly well. The first thing that happened was that every member of the class bonded with every other member of the class. We solved those problems as one giant team. We would band together in the student offices, in groups of three to five, and then representatives from each group would go wandering late at night from one office to the next, trying to see if one of the other groups had found a promising line of attack. All the experimentalists made friends with the theorists. All the theorists made friends with each other. Half of my best friends in grad school came out of that one course.
And because the lecture notes weren't comprehensible, we were forced to read the book very carefully, page by page. And the book happened to be Ashcroft and Mermin's Solid State Physics, which is one of the best textbooks I've ever known.
In the end, I got fairly good at solid state physics (for an experimentalist, anyway). I later signed up with an adviser who was well known for asking tricky questions about solid state during oral exams, but I was well prepared. My thesis ended up being all about topics that I learned in that class, and I even put the lecturer on my thesis committee, because he turned out to be a very nice guy and quite smart.