This quote, from Tim Daly, president of The New Teacher project, is the heart of the problem with tenure, I think:
Tenure says you can’t be dismissed unless you are shown to be incompetent through the evaluation process … but the evaluation process doesn’t work at all, so tenure is seen as an ironclad guarantee of a job.
One of the main reasons that process doesn’t work, at least in my experience, is that many of the people responsible for doing the evaluating know less about teaching and/or the subject matter than the ones being evaluated.
On the other hand, perhaps removing the protection of tenure, or extending the time required before tenure is possible, will have the effect of shifting the burden to the ones doing the evaluating. In other professions, an incompetent boss may get rid of valuable employees for bad reasons. However, if that behavior continues for long, either the company will fail because of its inferior work force or someone higher on the chain will correct or remove that incompetent boss. In either case, the blame falls on the leadership, not the employees. Could the same thing happen in education?
If so, perhaps we’d see a greater push on the part of leadership to identify the real factors contributing to the failure and to find ways of addressing them. Certainly, incompetent teachers are a problem, but incompetent teachers aren’t the only problem, and don’t even seem to be the largest problem. Instead, home life and socio-economic status seem to be the largest factors in predicting success. School resources and overall philosophy and approach to education are also significant.
But even if we restrict the focus to teacher incompetence, there are still two major hurdles to clear before we can start stripping away tenure and tying salary to performance: how we assess teacher performance, and who does the assessing.
Right now, the push is to tie teacher performance to a few standardized tests … mainly because the people responsible for evaluating teachers have no better ideas about how to do it. And that’s the second problem: a lack of leadership. In my opinion, there needs to be two branches of school leadership: one that deals with logistics and discipline, and one that deals with academics and instruction. Right now, these fairly different sets of priorities fall on the same shoulders, with the emphasis usually placed on the former.
I would suggest that the academics/instruction branch consist of experienced educators who have proven themselves (by whatever methods) to be effective teachers. These teachers would work side-by-side with the logistics/discipline branch to guide the school forward. However, teachers would be evaluated by and accountable to only the academics/instruction branch. In other words, the ones who have proven they know how to effectively educate children in a given subject matter are the ones guiding and evaluating teachers trying to reach the same level of skill. If teachers, over time, display a reluctance or inability to improve their teaching in a manner deemed appropriate to the panel of experienced educators, then they’re released.
This idea isn’t very new. It’s modeled on apprenticeship, and I’ve already seen it expressed to some degree through the National Writing Project (who’s motto is “teachers teaching teachers”) and the National Board for Professional Teaching Standards (which were developed by experienced teachers).
Errors and abuses could still occur in this model, of course, but I think it would be much more reliable than our current system because it would be based predominantly on proven educational methods and research. I suspect it would also engender less resentment from teachers and would provide a career step that didn’t involve leaving the classroom completely. It would also make the job of dealing with the logistical/disciplinary issues easier because those leaders wouldn’t have to worry about teacher performance and academics (as they currently do).
Finally, greater pay and professional respect would go a long way to improving the appeal of the profession (and therefore, improving the level of applicants), and a model that placed professional success and experience as a guiding force in the leadership of a school could command both.
I taught English and creative writing at a public high school. I stay involved in education through my work with the Northern Virginia Writing Project. 
Having worked in public schools for over 10 years, I would completely agree that administration, in its current structure, is spread far too thin to accurately and/or adequately evaluate teachers. The idea for separate “branches” of responsibilty is interesting…but hard to picture during a time when more and more curriculum directors are being cut to rescue budgets in crisis.