1) A teacher whose students consistently perform below grade level, but who jump 2 or 3 grade levels under her teaching (in the course of a year) is probably not a bad teacher. I know several teachers, particularly those who work with students for whom English is not a native language or who work with students with communiications barriers, where this is the case.
(My father is on the board of trustees of a school for the deaf and blind; this school consistently is one of the worst performing schools in the state, but at least part of that is because the vast majority of its students enter the school between 3 and 8 years behind grade level. I know the English teacher there, and she's responsible for some phenomenal gains, but even then it's rare that students will be reading and writing at grade level).
2) While we have a discussion of what bad teaching is, we also need to discuss what good teaching entails. "Sufficient progress on standardized tests" does not describe the teacher, but rather the students, and even then only on a single day. I'm willing to buy that there's some correlation between those tests and overall achievement, but would suggest that 3 hours is probably an insufficient amount of time to fully demonstrate one's learning over the previous 9 months.
Let's have a vibrant discussion of the how's and why's of teaching excellently, both in terms of what it looks like in action and what it accomplishes.
3) Finally, let's take a good look at what useful, worthwhile learning entails, and what it looks like. It may be that the system of bells and periods is more destructive than it is productive (I suspect this to be true, given my knowledge of my own workflow, as well as how I had to drastically alter my workflow when I taught), and it may be that we spend too much time on low-level skills (like rote memorization solely for the purpose of taking a test) and not enough time on internalization and contextualization of meaningful knowledge and skills (how many students have you known who could only solve a math problem when they knew ahead of time, by way of reading the chapter heading, what that problem would entail, but would not be able to solve that problem if they saw it mixed in with other types of problems?)
It's easy to say "Oh, let's fire the bottom 10% of teachers and replace them with excellent teachers", but such a statement falls flat on its face pretty quickly (for one, do we have a hidden reserve of these "excellent teachers", and if so, can we convince them to leave their more lucrative jobs elsewhere?).
It's much more difficult, but also much more worthwhile, to say "Let's figure out what excellent teaching and learning look like, in all their dimensions, and then let's see if we can figure out how to make it happen more often."
As far as education reform discussion topics go, I'd rather have the latter than the former.