I obviously don't know all the details of these two classrooms, and the data appears to show a real difference in the teaching abilities between these two teachers. However, let me throw out a few real scenarios that could provide other explanations...
If Teacher A's students get lower test scores and Teacher B's students get higher...
1) Teacher A specializes in working with lower level and learning disabled kids. He gets good results, although his students regularly don't make whatever the state deems "annual yearly progress" with his LD kids, so his results seem lower.
2) Teacher B is friends with the principal and is regularly assigned students who are already high performers. What, your boss never shows favoritism in your workplace?
3) Teacher A sees the standardized tests for the jokes they are, and concentrates on higher level skills that aren't measured well these tests - (processing, analysis, creativity, teamwork) all the while teaching the required reading, math, and science. Teacher B drills his students with the test prep books, the kids do ok on the tests, then forget everything. Teacher B's students do well as they continue on in middle school and high school because they have learned how to think, not just regurgitate. Teacher B's kids become part of the majority of High School students who can't really think, and whose scores and performance continually drop. Which class would you really like your child to be in?
Yes, I'm an elementary school teacher and no, I'm not just whining. Standardized tests are one measurement, but not the only or best one... just the cheapest and the easiest for politicians and lazy reporters to spout about. In evaluating teachers they should be considered by school administrators as one metric. The problem with what the LA Times has done is that while they say that there are other metrics for evaluating they present none.