Yesterday I wrote about Bill James' take on ground-ball pitchers, and I got so worked up that I forgot to mention something intriguing Bill wrote in the comments, in response to a comment about Derek Lowe perhaps having been ill-treated by the Indians last summer. Here's Bill:
Not exactly responding to CWright, but spinning off of his comment about Derek Lowe ... it is my thought in general, not merely with regard to ground ball pitchers, that we are missing an opportunity with regard to OLDER pitchers by being rigid and intolerant in regard to workloads. It just seems obvious to me (understanding that many things which seem obvious don't stand up to research, like the obvious advantages of ground ball pitchers) ... it just seems obvious to me that, as a pitcher ages, he may need more recovery time and/or may have slightly shorter limitations. But we basically say to older pitchers "if we're going to use you as a starting pitcher, we have to use you on the same starting schedule as we would use a 28-year-old, because we can't run different starting schedules for different pitchers," and we allow older starting pitchers to fail, rather than working around their limitations.
So I was thinking about this, and wondering to myself, "What would 'working around their limitations' look like?" One might argue that in the olden days, managers had a great deal more flexibility than they do now, because they constantly were adjusting on the fly. Teams played many doubleheaders during the season, and thus had more off-days than they have now. As recently as the 1930s and '40s, starting pitchers often started on just two or three days of rest, especially if they'd been knocked out early in their last start. Frequent rain-outs played havoc with a manager's projected "rotation" ... and in fact rotations as we know them didn't really exist until the 1960s.
A couple of examples ... Most fans probably figure that Lefty Grove, Earth's Best Pitcher in the 1930s, started every fourth game and threw a ton of innings. He didn't. Grove never led the American League in innings, and from 1926 through '33 -- during which he was quite healthy -- Grove averaged only 31 starts per season. He also pitched out of the bullpen from time to time. Most fans probably figure that Whitey Ford, the Yankees' ace in the 1950s, started every fourth game and threw a ton of innings. He didn't. In his healthy seasons, Ford averaged 30 starts.
That changed in 1961, when Ralph Houk replaced Casey Stengel as manager. Ford started every fourth game, led the American League in innings twice, and after a few years of that he broke down. I'm not saying Houk was responsible for the end of Ford's career; maybe it was just his time. My point is this: There was a time when managers, whether by preference or necessity, actually use their judgment regarding their starting pitchers' schedules.
That just isn't the case any more. You have your five starting pitchers, and by gorry each is going to start every fifth game unless there's an off day and then you might skip your fifth starter. But probably not. And so not a single major-league pitcher started more than 34 games last season.
Which maybe is a great thing. I don't know, and frankly I don't think the teams really know, either. Everybody's doing it the same way because everybody's doing it the same way. Great way to cover your ass. Maybe not the greatest way to win baseball games.
Enough digression. I was stalling because I don't know what it would look like, "working around their limitations". But it occurs to me that Bill's point also bears upon the Aroldis Chapman situation. Upper management doesn't (or didn't) want to necessarily
waste spend Chapman's considerable talents on just 65 innings per season. But in baseball these days, you're in one of two boxes: you're either going to pitch six or seven innings, or you're going to pitch one.
It's been decided that Chapman isn't ready to pitch six or seven innings. So he's going to pitch one inning.
Gee, guys. I'm not going to try to answer the question about 40-year-old starters. But here's something about Aroldis Chapman starters and Mike Leake starters.
The organizational goal this spring was to convert Chapman from a one-inning pitcher to a six-or-seven inning pitcher. In the space of about six weeks. This can work quickly and brilliantly (cf. Chris Sale). Often, it probably won't work so quickly or so brilliantly. The Reds have given up after five weeks. I'm not saying they're wrong, especially considering the personalities involved. But on paper, this seems the result of a stunning lack of creativity.
Why does Chapman have to throw six or seven innings now? Why isn't three or four innings good enough? Why not consider three or four innings just a way station on the way to five or six or seven? With Chapman back in the bullpen, Mike Leake's probably heading to the rotation. Mike Leake is a major-league starting pitcher ... but not by a lot. Might not he pitch more effectively if asked to throw just three innings at a time? Isn't it possible that 180 innings from Chapman and Leake is more valuable than 240 innings from Leake and Chapman? If distributed differently?
It's definitely possible. Maybe not all that likely. But possible.
I should mention that the Rockies did experiment last season with paired starter-reliever tandems; almost exactly what I'm suggesting the Reds might have tried. But in a sense, the Rockies were just as rigid as anyone else; instead of tailoring their plan to their individual pitchers, they went for the one-size-fits-all approach. Which was interesting but unsuccessful, and so they junked the plan after just a few months.
Which is more admirable, I guess, than refusing to come up with a different sort of plan at all.