It'd be interesting to see some stats for innings where the pitcher has been sitting in the dugout for say 25 minutes or more. We saw the same thing happen to Wright not long ago.
Koji was probably overused in April and May with a total of 21 IP. But in June and July he only had a total of 15 IP. That's very moderate usage. I'm not sure how you're supposed to plan around a reliever that you can only use sparingly with the number of innings that have to be covered. It's much easier said than done when you're trying to win games. IMHO.
A lot of unpredictable stuff happens in baseball. Nobody foresaw Wright pitching like this either. There are examples all over the place of guys suddenly finding something that makes them perform much better. It seems to be the nature of the game.