To nobody’s surprise, the Yankees have struggled to get significant innings out of their starters: in the first 18 games of the season, Yankees starters are averaging fewer than five innings per start, and have gone six or more innings just six times (three by Gerrit Cole, two by Jordan Montgomery, and one by Domingo Germán). Because of this, the relief corps has been used heavily — only the San Diego Padres and Tampa Bay Rays have seen their bullpens accrue more innings in the early going.
A solution could be piggybacking starters. Often utilized in spring training and the minor leagues in order to give two starting pitchers work on the same day, it involves planning for the starter to go only three-to-four innings by scheduling a long reliever to follow him. In a perfect world, these two could hypothetically finish the game; more typically, they combine to get through the sixth or seventh. Because neither pitcher should see any batter more than twice, two lesser-tier pitchers, properly used, can approximate the performance of an upper-middle rotation starter.
With all the question marks on the pitching staff, the Yankees are actually in a perfect position to employ this strategy, as recent years have seen them develop a small army of starter/reliever hybrids. In two appearances, Michael King has been electric, following six shutout innings in which he gave up only one hit on April 4th with three zeroes in his next outing twelve days later.
Although Nick Nelson has primarily been used as a one-to-two inning reliever this year, he was a starter in the minors; the same is true for Albert Abreu. Moreover, prospects Deivi García, Luis Medina, and Luis Gil could be eased into larger roles in this way. (Although both Medina and Gil may be a bit further from the Bronx, depending on the level the Yankees end up placing them when the minor league season starts.)
Part of the difficulty with this method is that you essentially have a spot in the bullpen that you only use once every fifth day — and possibly more, if you decide to piggyback multiple starters. So far, the Yankees have seemed to be averse to this strategy, as King has been the first to be demoted whenever a fresh arm is needed; by sending him down, he becomes unable to pitch for 10 days, making it impossible for him to serve as a piggyback.
King, however, is not the only member of the bullpen with options remaining: Nelson, Alberu, Jonathan Loaisiga, and Brooks Kriske all have options available or spent this season — and that’s just among relievers on the 40-man roster, which currently has a spot open! The Yankees actually have the option of keeping King on the active roster to pitch in relief every fifth day while still keeping the Scranton Shuttle operational.
Of course, it’s not quite as easy as this, and there are still many questions to be solved. Most notably, which starters should get the piggyback? The case can be made that any of Corey Kluber, Jameson Taillon, or Domingo Germán ought to be paired up with one, but only one can, realistically speaking. Do you bank on Kluber’s track record, and hope that he will cut down on the walks so that he can go deeper into games? Do you protect Taillon’s arm by limiting him to twice through the order? Do you simply partner the piggyback with Germán, objectively the starter with the shortest track record? If (once?) García takes a spot in the rotation, does he get paired up?
There are plenty of options that the Yankees could take. Personally, I would pair up King with Germán (well, I’d prefer García, but that doesn’t seem to be in the cards
due to service time shenanigans for some reason at the moment). At this juncture, it seems more likely to me that Kluber or Taillon could get deeper into games with a lead, and with wins at a premium at the moment, I’d prefer not to rely on King as a piggyback in those situations until he has a longer track record. No matter which option the Yankees pursue, however, the decision to piggyback starters could help keep the bullpen fresh for the long haul — if only they’re willing to take that step.