July 28, 2009

Pitching Workloads

Is 120 the New 100?

by Christina Kahrl

Printer-friendly

Contact Author

When Baseball Prospectus was first getting started more than a decade ago, we ended up helping add a certain pointedness to the question over what sort of workloads starting pitchers could handle by creating a statistic, Pitcher Abuse Points. Controversial at the time because of the inference that it was a predictive tool, what PAP was in fact was a counting stat, sort of like a warning light on the dashboard of your car-it gave you cause for concern, but it wasn't telling you that a complete breakdown was incipient. However, when PAP was first published, the baseball industry was perhaps better typified as relatively indifferent on the subject. Previous warnings about workloads had been aired-perhaps most famously in Craig Wright's and Tom House's The Diamond Appraised in the '80s-but generally speaking, pitcher workloads weren't as carefully monitored as they are now.

Fast-forward a decade, and that has changed almost industry-wide. Organizations monitor their pitchers from the highest level to the lowest, count pitches in games, on their throw days, warming up in the bullpen, even throws to the bases, if it involves a pitcher and a baseball in flight, it's being charted. In A-ball leagues, several teams use adaptive workloads with an eye towards keeping younger pitchers from being overworked-instead of a normal starter/reliever split in assignments, you'll find groups of pitchers pitching in tandem, paired off to handle the first six or seven innings together, and throwing 60 to 90 pitches, and trading off the honor of starting or following in that ballgame. As much as possible, teams are not sacrificing any of their pitching prospects for the greater glory of pennant race in Binghamton.

The extent to which there has been an overcorrection in terms of pitcher usage patterns is perhaps best reflected in the recent adaptations made by several teams in the last couple of seasons. Nolan Ryan's announced determination to make the Rangers' pitchers more durable hasn't led to 140-pitch starts or relievers on a pace to appear in 90 games. More teams are moving are realizing that the five-man rotation isn't so much an inflexible quintet as much as a five-day rotation in which you keep your four best rotation men working on their regular rest, and if the odd offday affords a team the opportunity to skip the fifth starter, then he gets skipped, or shipped to Triple-A to take a turn there until he's needed back on the big-league club. This isn't revolutionary or even evolutionary as usage patterns go-this is where we were 25 years ago. Similarly, the obsession with the nice roundedness of "100" has become less a line of death as much as a suggestion-as Rany Jazayerli noted in 2004, it's better to think of 120 pitches as the count at which people should really be concerned.

How much have matters changed? In 1999, the pitcher who averaged the highest pitch count per start on the season was Randy Johnson, with 120 pitches per game; he topped 130 pitches in a game eight times. Today, the per-game leader is Justin Verlander, with 109, which would have ranked 14^th in 2000; he's thrown over 120 pitches all of four times on the year.

Consider this list of the highest single-game pitch counts of the last five years:

Pitcher, Team               Date     Pitches
Livan Hernandez, Nationals  6/03/05   150
Livan Hernandez, Nationals  7/31/05   145
Livan Hernandez, Nationals  6/15/06   138
Tim Lincecum, Giants        9/13/08   138
Carlos Zambrano, Cubs       5/05/08   136
Livan Hernandez, Nationals  7/15/05   136
Aaron Harang, Reds          7/08/06   135
Carl Pavano, Yankees        5/17/05   133
Roy Halladay, Blue Jays     6/02/09   133
Curt Schilling, Red Sox     4/25/06   133

Livan Hernandez is and was a physical freak; the heaviest workloads in the industry didn't break him, but he'd also demonstrated the capacity to handle it. Roy Halladay is still with us. Tim Lincecum scared teams off in the draft because of his build, but he had demonstrated a remarkable ability to throw without getting sore in college, and his arm has not fallen off. On the other hand, Carlos Zambrano's had issues with his durability in recent years after establishing a brief reputation as a workhorse; he has also managed to avoid a complete breakdown. After his big ballgame, Carl Pavano subsequently became a watchword for fragility, breaking down later that season, and missing most of the next three.

The unhelpful suggestion is that not all pitchers are created equal. Pitchers like Kerry Wood-heavily overworked in high school-and Mark Prior might be the signature warnings of pushing any one pitcher too far since their breakdowns since 2003. Not pushing pitchers too hard before they've matured physically is generally well understood. The Dodgers won't push Clayton Kershaw too far down the stretch, just as the Rays have been careful with David Price, because they want to bank on the futures they'll get from those pitchers towards the tail end of their initial six years of service time. Clubs are actively protecting their investments in their talent. The general rule of thumb has been for pitchers to avoid getting worked too heavily before their age-24 season, but as Wood's experience reflects, there's only so much control teams have over younger pitchers. Because of service time considerations (not to mention overspecialization in bullpen roles), clubs are also reluctant to follow the old Earl Weaver rule of breaking in future starting pitchers in middle relief, instead targeting the age range in which they'll get those first six seasons before free agency with any premium prospect.

Similarly, there's a better understanding that not all pitch counts are created equal. Throwing 100 pitches in three innings is a lot more taxing than 100 pitches over seven-it's pretty obvious that kind of tally in that short a period of time means the guy's struggling, allowing baserunners, and dealing with the added stress of throwing from the stretch. That said, there's not necessarily a predictive element involved. Take the highest individual pitch counts in a single inning thrown this season: the 57 thrown by Francisco Liriano in the third on May 30, and the 56 thrown by Chris Young in the third on April 27; both pitchers made early exits, both made quality starts their next time out. Liriano's trying to salvage a season that started badly; Young's on the Disabled List.

A version of this story originally appeared on ESPN Insider .

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