February 11, 2003

Prospectus Feature

The Yankees' Seven-Man Rotation

by Nate Silver

Printer-friendly

Contact Author

As pitchers and catchers report to sunny climes this week--soon to be joined by hitters, beer vendors, and spring breakers--much will be made of the battle for the five slots in the New York Yankees' starting rotation.

The Yankees, you see, have seven handsomely--paid starters--what hubris!--any of whom could start on opening day for the Newark Bears or the Rancho Cucamonga Quakes, or if you give them a mulligan on Sterling Hitchcock, about half the teams in the major leagues. It is the greatest waste of talent, so it would seem, since Ocean's Eleven.

Here is our cast of seven, in most probable order of appearance:

• Roger Clemens
  PECOTA Projection: 26 GS, 168 IP, 3.56 ERA

  In an article that I wrote last month, I discussed how a 25-year-old pitcher is a more valuable commodity than a 22-year-old pitcher; the fact that the 25-year-old has survived the injury nexus intact provides a very important piece of information about his durability.

  The same isn't quite true when comparing a 40-year-old pitcher to a 35-year-old pitcher, but the difference between the two ages isn't nearly as great as you'd think, for much the same reason. Most pitchers don't reach age 35 with their careers still intact, much less age 40. But the ones that do have been chosen because of their superior durability. Given that Clemens has made it this far, and is still a near-dominant pitcher when he takes the mound (he struck out more than a batter an inning last year), his odds of continuing to be effective are pretty good.

  In a twist of fate that will warm hearts from Abilene to Wichita Falls, the Rocket's best comparable is Nolan Ryan, who won 71 games from his age-40 season onward, and struck out 1,437. Clemens is so wonderfully competitive that I can't see him hanging up his spikes until he absolutely needs to, and he has at least a fighting chance to become the first 350-win pitcher since Warren Spahn. Can the Advil commercials be far behind?

  PECOTA hedges its bets a little by predicting a further reduction in his innings pitched; a few skipped turns in the rotation are likely. But past the age of 38 or so, the risk of catastrophic injury to a pitcher is very low; can you name the last 40-year-old career starter who blew out his elbow?

• Mike Mussina
  PECOTA Projection: 29 GS, 195.3 IP, 3.69 ERA

  Mussina provides a good example of the danger of reading too much into a pitcher's ERA. Last year Mussina's walks were up just a hair, his strikeouts were down by about the same amount, and he was giving up home runs at just the wrong times; at the conclusion of play on August 11, Mussina's ERA was 4.95. A lot of folks were calling for his head.

  His ERA for the remainder of the season was 1.97.

  Mussina's strikeout-to-walk ratio is still outstanding and he has about as clean a bill of health as a pitcher can have. Six of his top eight comparables are Hall of Famers, and the other two are Jack Morris and Curt Schilling. Somebody--maybe Roy Halladay, or Matt Morris, or Tim Hudson--will exhibit a similarly high ERA, and the same panic will occur. ERA isn't particularly linear, and small, even random changes in underlying performance levels can lead to apparently large changes in results. The Moose will be fine, as he's always been.

• David Wells
  PECOTA Projection: 29 GS, 191.7 IP, 4.16 ERA

  Wells is fat, cranky, and has a bad back. For all of those reasons, he's pegged as a future BP author, as well as a career-ending injury waiting to happen. But appearances can be deceiving. Wells' season has been seriously derailed by injuries only once in the past eight years; Jim Palmer was an underwear model, and had his last good year at age 36.

  Although he might have trouble tucking in his jersey, Wells is an efficient pitcher who doesn't throw more pitches than he needs to. His approach is not entirely dissimilar to Tommy John and Jamie Moyer, his best left-handed comparables. The most serious risk he faces is that the Yankee defense, which hasn't been very good since 1998, will continue to erode as it ages.

• Andy Pettitte
  PECOTA Projection: 26 GS, 153.3 IP, 3.89 ERA

  In spite of just one trip to the DL last season, Pettitte made only 22 starts, and has the lowest IP projection of any of the big six starters. My colleague Will Carroll suggests that Pettitte's motion puts a lot of strain on his elbow, and that he is not the sort of pitcher who can be effective at less than 100%. If that's the case, then he could stand to benefit as much as anyone from the Yankees' abundance of starters. Depth doesn't merely provide a solution after injuries have occurred; it can also serve as a preventative mechanism by lessening the incentive to send a pitcher to the mound when he's not at full strength.

  Injury risk aside, Pettitte is an interesting pitcher. He's very tall for a lefty, and allows an absurdly small number of home runs; over the past three seasons, he's given up only about half as many homers as an average pitcher. PECOTA compares him to Kevin Brown and the ubiquitous Tommy John (as well as Jim Kaat). It's interesting to imagine him as a hybrid between two, combining John's refusal to do the hitter's work for him with Brown's disposition toward pitching hard and low in the strike zone.

• Jeff Weaver
  PECOTA Projection: 28 GS, 179 IP, 4.32 ERA

  Weaver's strikeout rate projects as slightly below league average. That's never a good thing for a 26-year-old, and his three best comparables (Luis Leal, Jaime Navarro, and Izzy Valdes) are about as discouraging as humanly possible. He could still have a long career since he does everything else pretty well, but his success could well be linked to his ability to keep the ball in the ground.

  Instead, Weaver went in the wrong direction after leaving Detroit last year. For all the talk about Comerica National Park and its (formerly) spacious outfield, Weaver's GB/FB ratio swung from 1.42 as a Tiger to 1.08 as a Yankee, and his home run rate went in the direction that you'd expect. He was pitching like a man who didn't trust his defense up the middle, and he was less effective, even after adjusting for park effects.

• Jose Contreras

  It would be dishonest to run a projection for him; statistical data is PECOTA's fuel, and there's none for it to work with, at least not until Clay Davenport comes out with Cuban League translations. What do we know about him (Contreras, not Davenport)?

  • An awful lot of teams were interested in him, and willing to pay a pretty penny in a down market.
  • He's done exceedingly well against international competition.
  • He's tall, has thick legs, and a big butt, and that's usually a good build for a pitcher.

  A lot of people are skeptical, recalling pitchers like Rene Arocha and Ariel Prieto who didn't live up to expectations, and others like the hermanos Hernandez who certainly haven't exceeded them. It's not clear that's a fair way to evaluate Contreras, any more so than predicting failure for Hideo Nomo would have been on the basis of Mac Suzuki's struggles early in his career. The more specific the information about Contreras, the more favorable it is. I'm a believer.

• Sterling Hitchcock
  PECOTA Projection: 9 GS, 56.7 IP, 5.18 ERA

  Hitchcock is the Art Garfunkel of the Yankees' pitchers, hanging on with his more talented counterparts only by a thread and an ill-advised contract. His strikeout rate is low, and he's one of the few pitchers whose hit rate allowed on balls in play has consistently been well above that of his teammates. Very few of his comparables rebounded to have productive careers, and those that did had to reinvent themselves as relievers.

Conclusion

If there's one theme that unites the Yankees' starting pitchers, it's that they're all, well, pitchers. Each one has at least three good pitches, each has good command, and each has responded well to changes in their physiology throughout their careers - at least so far. The downside is that none of them, Contreras included, is a particularly good candidate for the sort of two-pitch, one-batter-at-a-time mentality that successful relievers are normally associated with.

No, somebody will have to be the odd man out. The good news for Joe Torre is that, more often than not, the decision won't be in his hands. Here are the projected games started totals for the five pitchers that ended the year in the Yankee rotation last season:

Pitcher        PGS
Mike Mussina    29
David Wells     29
Jeff Weaver     28
Roger Clemens   26
Andy Pettitte   26
------------------
Subtotal       138

No pitcher in the Yankee rotation is a red light injury risk. But all pitchers are injury risks, and for various reasons, the Yankee hurlers are no exception. Only four teams last season had their top five pitchers account for at least 90% of their team's starts; we're projecting that the Yankees top five account for 138 starts, or about 85%. Which leaves just enough room for: 

Jose Contreras  24?
------------------
Total          162

It is unlikely to work out that conveniently, of course. Some of the injuries and missed starts will overlap; Sterling Hitchcock will get a handful of starts, and Danny Borrell or Brandon Claussen will ride the shuttle back and forth to Columbus at least once. But depth is a tremendous asset, especially when it comes to starting pitching, and if the Yankees understand that, it is another way in which they have distinguished themselves from the herd behind them.

Nate Silver is an author of Baseball Prospectus. 
Click here to see Nate's other articles. You can contact Nate by clicking here