It is a problem, but it's not the same problem. You can try to address the issue of secular trends in the industry by looking at say, performance relative to in

It is a problem, but it's not the same problem. You can try to address the issue of secular trends in the industry by looking at say, performance relative to industry peers. How you define industry peers of course is a bit tricky. Another option is to look at share price returns from the day before the new CEO was announced. That would incorporate the market pricing in the new CEO. However, I don't think this analysis did that.