No. It's actually quite common for the same teacher to be given lower-performing students year after year.
Since people can't seem to get this through their skulls: Students are not randomly assigned to classes.
My wife teaches history. Starting this year, her scores are going to go down and stay lower than all of the rest of the teachers. Is it because she suddenly started sucking this year? Or is it because she volunteered to take on a class of students who only learned to speak English in the last couple years? While the students are probably smarter on average than the rest of the school (as their parents worked very hard to live in an area with good schools), their test scores are invariably lower. They know how to speak English, but don't recognize all the vocabulary used on tests, and lack the cultural framework to understand other things. She can't even show improvement with them because they don't have any previous tests to compare to.
Also, a few years ago, a new history teacher was hired. She turned in the highest scores for the grade. Is it because she was the best teacher? Was it just luck? Or was it that the administration decided to give the majority of the students on special education plans to the other two teachers? I know people think that was wrong, and that she should have been held accountable for the fact that she couldn't handle it. That's fantastic, as yet again, we've lost sight of the actual goal here. Those students were moved in order to ensure they got a better education. Yes, the other teachers could have complained about the fact that their scores were going to go down, but they didn't because unlike the majority of people pushing for test scores as evaluation metrics, they actually care about the students.
And I can already see the argument: "Well, those shouldn't be considered for the metric..." Maybe not, but federal laws say that those students' scores cannot be treated any differently than any other student. This, in the end, is the biggest difference in quality between private schools and public schools: private schools hide the scores of their special education students -- or simply expel or refuse to admit them. Most of the charter schools and private schools in this area simply won't take special needs students. When they do, they put them in special programs, which basically means: The same classes but with a tutor paid for by the parents and a stipulation that the student's scores aren't part of the rest of the school's statistics. Oh, and they usually make the county pay for the psych and disability evaluations (~$40K a kid, if I remember right)... because they can do that.
You see, when you have zero experience in this, it's easy to say "Well, a mediocre system is better than nothing", but that's just because you don't know what you're talking about and you can't see that the variance in scores among groups of students is greater than the variance produced by the quality of teacher. This means that no matter how you try to do it, if you base your metric on student performance, what you are measuring is trends in student grouping, not teacher quality.