Chronicles of the Lads

EARL WEAVER WAS RIGHT
Scioscia says that deciding between Ramon Ortiz and Aaron Sele is "by far, this is the toughest decision we've had in five years here." Given their performances in 2003 and this spring, this decision looks like one between arsenic and anthrax. Whither the four-man rotation? Every spring, we have 25+ teams trying to figure out which of their marginal starters will suck least and win the coveted fifth spot.

How did fifth starters do last year? To find out, I went to baseball-reference.com, and simply looked at the pitchers that had the fifth-most starts for his team (in some cases, like Casey Fossum and Jeff Suppan in Boston, I combined two pitchers or more, as they combined to be the fifth starter; I also excluded players like Brian Anderson or Sidney Ponson, who were traded, and Randy Johnson, who was injured). The final totals will include relief innings, and will not include everyone who made a start as a fifth starter. I think that’s okay; I’m just trying to determine what the general quality of a pitcher trusted with every fifth start. Also, some of the guys who get four or five starts are ostensibly worse than the guys I included.

Incidentally, did you know that Seattle had only five men start games for all of last year? I picked Gil Meche as the fifth starter, as he had the least innings.

Anyway, I identified 39 pitchers, and this is what I found: over the course of the season, they went 192-261 with an ERA of 5.18 and a RA of 5.57. Egads.

What is the point of running out a pitcher like this? The benefit is that the rest of your guys can pitch on four days’ rest instead of three. Is that a benefit? There’s no real evidence that pitchers are better off with four days’ rest. Does having a fifth starter save the bullpen? Let’s look at this in theory.

Let’s say we have four pitchers who average six innings a game and one that averages five. (Last year the top four Angel starters [which includes Sele, who was oft held to a 5-inning limit] averaged 5.8 innings per start, while Kevin Appier averaged 4.9.) Let’s also say that a team needs to fill up 1,458 innings in a year (162 x 9, and hoping that extra innings and ninth innings not pitched on the road essentially even out; last year, the Angels threw 1,431 1/3 innings, so that’s close enough).

In a five-man rotation, two starters get 33 starts and three get 32. That means 780 innings go to the top four guys and 160 go to the fifth guy. That leaves 518 innings to the bullpen.

In a four-man rotation, two starters get 41 starts and two get 40. That means a staggering 972 innings go to the starters and 486 go to the bullpen. 160 innings that used by a pitcher with a 5.18 ERA are gone, and so are 32 innings pitched by relievers – the back of the bullpen – and replaced by innings pitched by the good four starters.

But, you object, wouldn’t a four-man rotation mean that the starters pitch less innings per game? Quite possibly. Let’s say that we keep a four-man rotation, but assign the relievers 518 innings (which they had in the five-man scenario). That leaves 940 innings to the top four starters: 5.8 per game.

Of course, in practice, players get hurt, they get tired, they get traded, so the actual numbers wouldn’t come out exactly as I’ve outlined. But in my reduced-starter-inning four-man rotations scenario, the four good starters account for 64.5% of the innings pitched; last season, the Angels’ top four starters accounted for 49.8% of the innings pitched. Even with the top four starters pitching less innings per start, they still account for a higher percentage of the innings that the team needs pitched.

Think about the further benefits of having only four starters. For one, you get an extra spot in the bullpen – one of which can be a spot starter. Best of all, you can carry an extra position player: a hitter for late-inning pinch duties, a pair of legs, a reliable glove.

In my world, both Ortiz and Sele would be gone, and Shields would be the swingman/spot starter.

This will never happen.