On Monday afternoon, Astros manager A.J. Hinch made a move that—had things gone a little differently—would have echoed deep into the offseason. With one on and one out in the fifth inning of ALDS Game 4 against the Red Sox, Hinch called out to his bullpen and asked for Justin Verlander to make his first professional relief appearance. Ever. It was a strange move because, unlike the Red Sox, who had already inserted their ace Chris Sale into the game in relief, the Astros were not facing elimination in the game. On top of that, Verlander figured to be the Astros’ starter for Game 5, if it was needed. Why was Hinch making a desperation move?
Immediately after entering the game, Verlander gave up a home run to the first batter he saw, Andrew Benintendi, and what had been a 2-1 Astros lead turned into a 3-2 Red Sox lead. Let’s just say that for a few moments, Hinch was not a popular person among the citizenry of Houston. In hindsight, we know it turned out all right, but for a while, it looked pretty grim.
The Verlander move was the correct decision. It would have been the correct decision even if the Astros had lost and were facing Game 5 tonight without their ace, and understanding why is the key to understanding everything you need to know about the playoffs and #bullpenning.
There’s a case to be made that if Hinch were going to use Verlander anyway, he should have used him to start an inning fresh, with the criticism being that Verlander, as a starter, has never entered into a game mid-inning. Maybe that argument does hold some water, but the the majority of relief appearances involve a pitcher coming into a “clean” inning as well. While relievers would have some experience entering the middle of the fray, even they only do it sometimes.
But why go to Verlander at all when the Astros had a perfectly good set of options already in the bullpen, with relievers Ken Giles, Chris Devenski, and Luke Gregerson all finishing in MLB's top 50 among pitchers who logged at least 50 innings?
Let’s for a moment consider the situation in which Hinch found himself. He could use Verlander, who had pitched four days earlier, essentially for a short-rest start. Perhaps he could get 3-4 innings out of him right then. He could also save Verlander for a potential Game 5, in which he would hope to get six or seven innings from him. Maybe Game 5 wouldn’t even be necessary. But why short-circuit Game 5 when it was a distinct possibility to happen?
Well, maybe the situation wasn’t as bad as people initially thought. Game 2 starter Dallas Kuechel would have been available on full rest for the Astros in Game 5, and Boston had already inserted Sale into the game (he would go on to throw into his fifth inning of relief), meaning that if the Red Sox had countered with their Game 2 starter, it would have been Drew Pomeranz, and probably some succession of Doug Fister, Joe Kelly, and then checking to see if David Price still had anything left in the tank after throwing a combined 6 2/3 innings in Games 2 and 3.
If the Astros had burned Verlander and still had to play Game 5, Hinch already knew that they’d be facing a team with some very limited pitching options. In other words, Hinch had some wiggle room. He could afford to lose this game, both for the fact that there was a “tomorrow” (or more accurately, a “two days from now”), which would be played in Houston, and his pitching staff would likely be fresher than John Farrell’s. Game 5 would have eventually devolved into a bullpen game anyway, but the Red Sox would have started #bullpenning earlier than the Astros, and that can make a huge difference.
But with all that said, even without the knowledge that the Red Sox had burned Sale or that they were going to be reliant on Pomeranz and Fister, the Verlander move still makes sense.
The reason why lies in the game situation at the time. Hinch knew that he could have 3-4 innings of Verlander today or six innings in two days (if necessary). In the playoffs, trading six innings of coverage for three is an easier decision to make because of the number of off days and the (related) ability to have a high-end reliever like Giles pitch multiple innings, if needed, to make up the difference. In the regular season, burning through relievers like that and having to make up three innings can pile stress onto a team’s pitching staff. In the playoffs, that doesn’t matter as much.
But at the time that Hinch made his decision, he knew something else about the innings that Verlander could end up pitching. The Astros had a small lead and it was the fifth inning. While it’s hard to have a super-high-leverage situation in the fifth inning, a one-run lead for the pitching team is the score that has the highest leverage at any given point in a game. The two most important runs in a baseball game are the run that ties the game and the run that unties it, and with a one-run lead, a pitching team is on the precipice of giving up both. (Little did Hinch hope that Verlander would jump off that cliff in his first batter.)
Hinch, at the time of his decision, could bring Verlander into a close game that was relatively late today or have him pitch in a Game 5 that might have been a nail-biter or might have been a blowout. Hinch knew that he had a bird in the hand right there, and so he went for it. The point of strategic decision-making is not to set a strategy that keeps your hopes of winning alive for the longest amount of time. It’s to mass your resources into the situations where they will do the most good, especially in a short series where a win is exponentially valuable.
In Hinch’s head, he knew that he had a big opportunity with already having the lead. He probably had some sense, whether mathematical or not, that this was likely to be one of the biggest opportunities his Astros would have. And he wanted his best.
The only issue left to probate is whether Verlander represented his best option, compared to calling on Devenski or Will Harris. That one is harder to fathom. Verlander had never relieved before, so there was no telling whether he'd take to it. (It turned out that, other than the Benintendi homer, he threw 2 2/3 scoreless frames.) He likely would be able to give the Astros multiple innings, meaning that whatever he gave them would probably replace the combined work of two or three of the “usual” bullpen arms (and they could still be called on if needed).
I don’t know whether Verlander really was mathematically the best option right there, compared to Devenski/Harris/Gregerson, once everything is factored in, but I feel confident that Hinch felt it was his best option and probably one that the manager had already planned in his head the night before. So let’s look at Hinch’s decision tree. At the moment he made the decision, he knew that starter Charlie Morton needed to hit the showers, and he needed something good out of the bullpen.
First, the negative outcomes:
Go with Verlander, and the Astros lose the game: You still have Kuechel for Game 5 (against a tattered Red Sox pitching staff), while saving some arms from working, meaning that if Game 5 devolves into a bullpen game, the bullpen is more fresh than it would have been. Verlander can pitch Game 2 of the ALCS if the Astros win Game 5.
Go with the “real” relievers, and the Astros lose the game: Verlander pitches in Game 5, while the bullpen is a little more tired (although Kuechel could probably play Verlander’s relief role). Verlander would probably not pitch until Game 3 of the ALCS, if the Astros got there.
Verdict: Looking at the two circumstances where the Astros lose Game 4, the fringe benefits of Verlander pitching are slightly nicer.
Now, on the other side:
Go with Verlander, and the Astros win the game: The series is over, and Verlander can pitch Game 2 of the ALCS.
Go with the “real” relievers, and the Astros win the game: The series is over, and Verlander can pitch Game 1 of the ALCS.
Verdict: That’s really the only difference if the Astros win. So, by bringing in Velander to Game 4, Hinch is hedging his bets a little. He gives up the chance of Verlander pitching in Game 1, but doesn’t run the risk of him having to be pushed all the way back to Game 3. In return, Hinch gets what he clearly perceives to be his best option into a tight playoff game, and the circumstances are that he has some wiggle room. All he has to be willing to do is never hear the end of it if it backfires.
For being able to make that call, despite that danger, I salute him. The fact that Verlander gave up that home run, and even that he might have ended up losing the game, don’t matter. You have to judge a decision like that based on the knowledge that a manager had when he made it. It’s not that Verlander was miles away the better option for Hinch. But he was probably the somewhat better available option, and so it made sense to bring him in, because this was a high-value game, and every little bit counts.
And so that brings us to bullpenning, which has become the new fun-word in baseball. The idea that everyone is now a reliever is all the rage. In some sense it’s true, although like “the Andrew Miller revolution,” don’t expect it during the regular season any time soon. This is the sort of thing that works because of the off days, plus the fact that you don’t need five (or in some cases, four) starters and the good relievers can go longer.
Managers have (always) realized that leads are precious and in the context where you can only lose two games (and only need to win three) the stakes are even higher. They have new options for protecting those leads. Why not use your best pitchers in the situations that call for them and give the most delicate situations to your best pitchers, even if it would look a little weird (and unsustainable) in the regular season?
This isn’t the regular season.