Dave Roberts managed Game 5 of the NLDS like a college coach, down to not starting the youngster, calling for the idiotic sacrifice bunt, leaving his closer in 10 pitches after he was too tired to throw strikes, and using his two best starters on short rest. But it makes sense: College teams, unlike their professional counterparts, usually have, at most, half a dozen trustworthy pitchers. I say “most” because the Dodgers — despite trading for the best starter moved at the deadline, making a big international signing, calling up at worst the second-best pitching prospect in baseball, and spending tens of millions in domestic free agency — still ran out of pitchers.
So Roberts and Cleveland’s Terry Francona — whose no. 2, no. 3, and no. 4 starters have all gotten hurt at the worst possible time — are being forced into throwback pitcher-usage patterns out of necessity. Relievers are already being used more than ever, which is one way to increase the impact of a team’s best per-inning pitchers. The other way is to start your best pitchers on short rest.
As recently as 2009, the Yankees went to a three-man playoff rotation, but in the era of two wild cards per league, there have been only 18 instances of a pitcher making a playoff start on three days of rest or fewer, and four of those have come in the past week.
So that brings up two questions: Are pitchers starting on short rest more often nowadays? And should they?
Ben Lindbergh and I discussed the idea briefly on The Ringer MLB Show Tuesday, but let’s take a closer look.
2016, After Relief Appearance: Clayton Kershaw, Game 2 of NLCS
2016, After Start: Clayton Kershaw, Game 4 of NLDS; Rich Hill, Game 5 of NLDS; Corey Kluber, Game 4 of ALCS
2015, After Relief Appearance: Noah Syndergaard, Game 2 of NLCS; Chris Young, Game 4 of World Series
2015, After Start: Dallas Keuchel, AL Wild Card Game; Clayton Kershaw, Game 4 of NLDS; Yordano Ventura, Game 4 of ALDS; John Lackey, Game 4 of NLDS
2014, After Relief Appearance: Yordano Ventura, Game 2 of ALDS
2014, After Start: Clayton Kershaw, Game 4 of ALDS
2013, After Relief Appearance: Lance Lynn, Game 4 of NLCS; John Lackey, Game 6 of World Series
2013, After Start: Clayton Kershaw, Game 4 of NLDS
2012, After Relief Appearance: Lance Lynn, Game 1 of NLCS; Tim Lincecum, Game 4 of NLCS
2012, After Start: Hiroki Kuroda, Game 2 of ALCS
The first thing that jumps out is that every short-rest start isn’t created equal — not only do they vary in length, but sometimes the last outing wasn’t an all-out start, but a brief relief appearance on a throw day. The second thing that jumps out is that Kershaw just blows up the whole sample. He’s got four of the 10 true short-rest starts on that list — plus his last start, in which he went seven innings in his fourth appearance in 10 days.
Pitchers started on short rest six times in 2015, and four times so far this year. With a round and a half of games still remaining, and the leaders in both championship series making up their rotations as they go, it’s not out of the question that we’ll hit six again (if not more) in 2016.
So should teams be doing this more? Well, teams that start a pitcher on short rest are 13–5 since 2012, so that’s a good sign, albeit one in an extremely small sample. But it’s about more than winning.
In amateur ball, this is less a strategic question than an ethical one. If a college coach used his ace the way Roberts has used Kershaw, my soul would leak out of my ears and go out into the world in search of bloody vengeance, but pro baseball is different. First of all, a college coach overusing a pitcher exposes a player to injury, which at best subjects an unpaid kid to unnecessary risk of injury and at worst could cost him millions down the road. Kershaw’s in the middle of a $215 million contract; if he blows out, his grandkids won’t be any less rich. And as a unionized adult worker whose employer has a top-notch medical staff, Kershaw’s better protected than a college kid.
In short, a big league manager (1) has a much better idea of where his pitchers’ limits are, (2) would never dream of throwing Kershaw for 150 pitches on three days’ rest, and (3) probably wouldn’t get away with it if he wanted to. That, and the fact that Kershaw and Kluber have a history of holding up over a 200-plus-inning workload, gives Roberts and Francona some ethical wiggle room.
That Kluber and Kershaw are making these starts is instructive, because the pitcher starting on short rest will always be diminished in some way. He won’t be as sharp, or will tire more easily, or won’t throw as hard, so not only does he have to be better than the alternative, he has to be a lot better. Jake Arrieta is starting on regular rest because the Cubs have a deep enough rotation to avoid any temptation to use him that way.
The problem is there aren’t that many guys who are both good enough to make this worth trying and don’t have any major injury or workload concerns. The Dodgers wouldn’t do this with Urias, nor — I imagine — would the Rangers with Yu Darvish, who’s just back this year from Tommy John.
Maybe, then, other teams should look to Kershaw’s closer performance in Game 5 of the NLDS for inspiration, not Game 4. Or they could look at Madison Bumgarner in Game 7 of the 2014 World Series for a more famous example. Instead of using short rest starts to shorten the rotation, managers could more liberally use starters in relief on their throw day to shorten the bench in the bullpen.
Mere days before Bumgarner pitched his way into legend in 2014, Sam Miller wrote about this tactic at Fox Sports, and he found that starters on their throw day performed quite well out of the bullpen. (The same small-sample-size caveats apply here, too). Swapping out a scheduled bullpen session for a relief appearance would be less of an added strain than moving a start up a day, and it opens up younger pitchers as weapons in relief.
When that article came out, I raised some objection I can’t remember to Sam, and he pressed me on it — why did I think he was wrong? The truth is, I didn’t want his conclusion to be true, because the idea of Bumgarner starting Game 1, closing Game 3, and starting Game 5 just seemed brutally unfair, which is another way of saying I thought it would work.
Despite all the work Kershaw and Kluber have put in this postseason, managers still probably aren’t getting as much out of their best starters in crunch time as they could. Nevertheless, pushing for more is a risk even under ideal circumstances; overuse can lead to Tommy John surgery or worse. So it should take a dire emergency for a starter to pitch on short rest or come out of the bullpen. But as bullpen usage starts to reflect that every playoff game is an emergency, perhaps rotations across the league will begin to do the same.