Thinking about how relievers often alternate good and bad years, I've wondered if there is a better way to use relievers. Phrasing it differently, I've wondered if bullpen construction/usage is a future market inefficiency. I liked this quote from the story: "Imagine if every team in the NFL used the same 3-4 defense. That's essentially what is happening in baseball. Everybody runs their bullpen and their pitch count policies the same way."
I found the stats in the article interesting, but they didn't suggest what should be changed. Getting away from the defined role of 8th inning/setup man and closer-only-in-save-situations is only part of the issue, in my opinion. I'd like to see a team make fewer pitching changes, and allow more relievers to work 2 or 3 innings at a time - because I believe warming up for only 1 or 2 batters, and making 70+ appearances a year causes too much physical wear and tear.
What if a team had 2 of relievers who threw 2-3 innings at a time, twice a week. Averaging 5 innings a week would equal 130 for the season. With 2 relievers on this schedule a team could...
...get through a game using 1 reliever more often, giving the rest of the 'pen an entire night off.
...develop future starters in the majors, since they'd probably use 3 or 4 different pitches rather than 2 like many relievers.
...carry fewer pitchers on the active roster (11?), which benefits the offense/defense with a deeper bench.
Going back a few years, I think this type of usage might be ideal for a guy like Rich Harden, who could dominate but was not durable. Scott Baker may have been another possibility.
</rant>