clock menu more-arrow no yes mobile

Filed under:

Joe Girardi's new-old bullpen philosophy

The Yankees have an unusual bullpen. Due to their collection of pitchers like Chan Ho Park, Joba Chamberlain, Alfredo Aceves, and Sergio Mitre, who are not only viable starters but conditioned to be starters, Joe Girardi can, if he so chooses, dispense with the match-up-based relief tactics that have come to dominate bullpen strategy in the age of Tony LaRussa and reinvent the long-man -- a pitcher who simply throws a few relief innings instead of jogging in and out to face one or two batters. These pitchers had seemed to be ticketed for extinction, having been outcompeted by specialists, but whether by plan or by accident, the Yankees are well-positioned to ignore this frequently counterproductive strategy and just let their best relievers pitch for as long as they can.

Rather than rush to mint an "eighth-inning guy," as the media seems very anxious for him to do, Girardi can instead create something of a shadow rotation of pitchers who can go two-to-three innings a night. As Chan Ho Park demonstrated yesterday, when it works, the long-man eliminates the need for the designated eighth-inning guy and the spot reliever. Rather than burn half your bullpen in the space of an inning or two, you let the guy rip until you’re ready to hand the game off to Mariano Rivera. The possibility of the same thing happening was there on Tuesday night.

This is the brave old world of reliever usage, one that was dispensed with for no good reason. It used to be very common for relievers to throw 100 innings a year. In 1982, to pick a year at random, 21 pitchers pitched 100 innings without making a start, which then equated to almost one per team. Among this strategy's benefits was that the team’s best relievers worked far more often than its worst, who might not even be in the majors because with one reliever absorbing so much work the staffs could be smaller. With the larger staffs of today and the influence on one-batter match-ups, there is little opportunity for a reliever to work so often. The last reliever over the 100-inning barrier was the Yankees’ Scott Proctor, who threw 102.1 innings for Joe Torre back in 2006.

Proctor paid a price for that workload, which suggests a couple of things: not all pitchers are equal to that workload and managers should only assign that much work to select pitchers, or that the top innings threshold for an ideal reliever should be somewhat lower than 100, or that appearances must be spaced out. Proctor made 55 appearances that year on no or one day’s rest, and also threw more than an inning on 37 occasions. The multi-inning reliever might need to be handled more sympathetically. For the Yankees, who have four such pitchers, this might not be a problem.

That said, it does not follow that the 100-inning workload is automatically injurious. Using the same 100-inning, zero start threshold I mentioned earlier gives us a population of roughly 350 seasons to choose from. There are many pitchers who appear on the list multiple times, for example Bill Campbell (five times), Bob Stanley (four), Bruce Sutter (five), Dale Murray (five), Dan Quisenberry (five), Duane Ward (five), Gary Lavelle (five), Gene Garber (six), Goose Gossage (four), Greg Harris (five), Hoyt Wilhelm (six), Jack Baldschun (four), Kent Tekulve (six), Mike Marshall (five), Pedro Borbon (four), Ron Perranoski (seven), Scott Sullivan (four), Sparky Lyle (six), and Stu Miller (four). If handled appropriately, relievers can do this without killing themselves.

I don’t know how far outside of the box Girardi is willing to think, or more accurately, can he get so far outside of the box that he’s back inside of it again, back inside the old box. The fact is, though, that Aceves, Chamberlain, Park, and perhaps (suspending disbelief) Mitre are overqualified to be one-inning relievers.

Baseball is going to have to go this way in the future, so the Yankees might as well be the first. With relievers taking on a larger workload with each passing year (AL relievers’ encroachment on starters has been steady but fractional over the last five years; last season they consumed 34.6 percent of possible innings), it is only a matter of time before some teams realize that there aren’t enough good relievers to go around and the solution has to be not more relievers, but exploiting the few good ones more thoroughly.

The Yankees, assuming this season is consistent with the recent past, will need about 520 innings out of their relievers. Mariano Rivera will get about 65 of them. If the four long-men can pitch 90 innings each, that leaves only 95 innings that have to go to Damaso Marte, David Robertson, Boone Logan, Royce Ring, and whoever else the Yankees will be forced to call up over the course of the season. The 13th reliever can stay in Scranton where he belongs. The 12th-best pitcher (and he may turn out to be Mitre, not Robertson), can watch most of the games from the best seat in the house, getting only trash-time work in games that are decided.

For a long time, baseball has done things the opposite way, with far too many mediocre pitchers getting to do important work in the bullpen. Here’s Joe Girardi’s case to make a bigger mark on the game than winning a second championship for the Yankees but to change the dominant pitching philosophy of the game.


  • Dead Player of the Day #9 is up, with Brooklyn Dodger Karl Spooner and existential questions about the Pittsburgh Pirates