At Home Plate - Whither Hast Thou Gone, Four Man Rotation? (Part I)

Whither Hast Thou Gone, Four Man Rotation? (Part I)

Written by Justin Zeth (Contact & Archive) on May 15, 2007

What makes a five-man rotation better than a four-man rotation? Wait; don't answer that just yet. First let's answer the corresponding question: What makes a four-man rotation better than a five-man rotation?

Well, for an easy example, let's consider the 2007 Minnesota Twins. The 2007 Minnesota Twins' five starters, primarily, are: Johan Santana, Boof Bonser, Ramon Ortiz, Carlos Silva, and, in somebody's idea of a practical joke, Sidney Ponson (That's the rotation at least until Terry Ryan puts down the crack pipe and gets Matt Garza and Kevin Slowey the hell into Twins uniforms.)

The illustration from there is obvious, right? Which is better:

8 starts from Johan Santana, 8 starts from Boof Bonser, 8 starts from Ramon Ortiz, and 8 starts from Carlos Silva; OR
32 starts from Sidney Ponson.

That's obvious, right? If I'm going to take 8 starts away from Johan Santana and give them to Sidney Ponson – ignoring for now the other 24 starts involved – I had better have a very good reason, and I'm talking a reason like “Santana got drunk one night on the wrong side of San Francisco and now has a voodoo curse on him that will cause his arm to explode the day he makes his 35th start, not including playoffs, in a calendar year.” Now, tell me, aside from that, what good reason is there? Why, that too is obvious. Think real hard about this, and let me know if I'm missing anything, but I'm pretty sure there are two, and only two, possible explanations for why I would want to take 8 starts away from Johan Santana and give them to Sidney Ponson:

1. It helps prevent pitchers from injuring themselves.
2. Resting starters an extra day makes them more effective when they do pitch.

Reason #1 is really the more prominent of the two, but it's also less interesting, so we'll give it a sentence or two later. I want to talk about reason #2 there for a minute. There are two primary rebuttals to the “more rest makes pitchers better” logic:

1. Five-man rotations came into vogue in the mid-1980s and became dominant in the early 1990s. Take a cursory glance at pitching numbers leaguewide over that time period and tell me: Does it look to you like the pitchers are pitching better than they were in the 1960s or 1970s?
2. Even if the extra rest DID make Johan Santana a better pitcher, it's dubious at best to assume it makes him so much better that it's worth taking 8 starts (which is about 50-55 innings) away from him and giving it to a lesser pitcher.

To put some perspective on that, which would you rather have: A pitcher who throws 200 innings with a 3.40 ERA, or a pitcher who throws 250 innings with a 3.65 ERA? Unless you have a full house of three Trevor Hoffmans and two Scot Shieldses in your bullpen, you want the guy who throws a lot of innings.

So there's little evidence that four days of rest instead of three makes J. Random Ace a significantly better pitcher, and even if it does, that doesn't mitigate the lost innings. There are exceptional cases – fortysomethings like Roger Clemens and Greg Maddux wouldn't do so hot if you suddenly threw them into a four-man rotation – but we need a better reason than “it makes the pitchers better” to take 8 starts away from Johan Santana.

That brings us to the real reason the five-man rotation is universal in modern baseball: We all believe it keeps pitchers healthier than they used to be. And I'm here to say, it just ain't so. First, let's observe that if a four-man rotation were as destructive as modern thought would have it seem, then we'd expect pre-1980 history to be littered with many, many more ruined arms and very few guys who pitched a high level for a long time. We find no such thing; in fact, we find that pitchers' injury rates now, 20 years into the five-man rotation era, are not substantially different from their injury rates in the 100 years of history before. Pitchers have very high attrition rates; a high percentage of them blow out their arms. This was true in 1890, when the radical new idea was the three-man rotation (instead of just two), and it was true in 1920, when four-man rotations were taking hold, and it was true in 1950, when pitchers everywhere were handing out walks like so many Pez candies and racking up obscene pitch counts, and it was true in 1975, when you needed 300 innings pitched just to make the league top ten, and it's true today, when 34 starts leads the league (and we wonder why the 30 win pitcher is dead.)

Bill James pointed this out years ago, but if you think pitching often and pitching lots of innings ruins pitchers' arms, just look at the 1970s. Numerous guys blew out their arms in the 1970s, but there were also many guys who didn't. Look up the innings-pitched counts for Steve Carlton. Nolan Ryan. Mickey Lolich. Gaylord Perry. Did pitching 320 innings every year destroy these guys' arms? Look at Phil Niekro, and Tom Seaver. Everyone was doing this. If pitching that many innings were destructive, we wouldn't expect to see anywhere near that number of Hall of Fame-level pitchers, especially not with some of the longest careers anyone's ever had.

Baseball is different now, though, and I can't ignore that point. Pitchers are throwing more pitches than they used to, not a lot more, but a little more, and that makes a significant difference on the back end; a 125 pitch outing, say Rany Jayezerli and others, is twice as hard on a pitcher's arm as a 115 pitch outing. But the more important difference between baseball now and baseball before now is how much longer the games take to play. Average game time is up nearly a half-hour from what it was 25 years ago. There are various reasons for this, which, in order of importance, probably are: 1. Batters stepping out after every pitch; 2. Managers changing pitchers as often as Jim Leyland changes cigarettes; 3. Increased offense; 4. More pitchers taking their sweet old time even after the batter has finally saved $200 on his car insurance and gotten into the box.

You might argue that (2) above is really just a consequence of (3); I might argue that it may actually be the other way around. Baseball's typically hare-brained response to this problem was twofold: 1. Put a timer on pitchers. 2. Don't enforce it.

So then, is the five-man rotation made necessary by the nuances of modern baseball and its tendency to move as fast as David Ortiz running underwater? Or is it a strategic move that could, conceivably, be reversed by, well, not a forward-thinking, but more of a “Back To The Future” kind of front office? We'll talk more about that... after this word from our sponsors!

Add this page to your favorite Social Bookmarking websites