Only decent, and only proxy. Say you worked the first 5 years propping up a dying business, and the next 5 years building a bunch of incredible value. The stock market may not have reacted to that, but your company is set to explode in value. Or you may have started in a bubble, or finished in a bubble, or...
So of course it is not the only measure, but it is certainly an extremely large part of measuring a CEO, since it is his job to maintain and/or grow the value of the company. It will never tell you if the company is bettering humanity, good to its employees, or other very worthwhile measures.
So why are you completely ignoring dividends and a stock split?
Looking at 10 years of stock performance while ignoring those factors isn't decent analysis, it's terrible.
I would argue that by other measures, Microsoft under Ballmer has not seen improvement either. These might include market power (the power to dictate future market directions), product sales share, product mind share, cultural association, recruiting, innovation, etc.
That said, I think there are a few areas where Microsoft has seen improvement under Ballmer. Xbox went from a "what? really?" product, to the clear console leader. And Microsoft government affairs is miles ahead of where it was under Bill Gates--from its own federal conviction, to instigating multiple actions against competitors.